Support multi-res occ grid & prop net (#176)

* multres grid

* prop

* benchmark with prop and occ

* benchmark blender with weight_decay

* docs

* bump version

.gitignore

@@ -119,3 +119,4 @@ venv.bak/
 
 benchmarks/
 outputs/
+data
\ No newline at end of file

docs/source/apis/generated/nerfacc.ray_resampling.rst

-nerfacc.ray\_resampling
-=======================
-
-.. currentmodule:: nerfacc
-
-.. autofunction:: ray_resampling
\ No newline at end of file

docs/source/apis/utils.rst

@@ -17,7 +17,6 @@ Utils
    render_weight_from_alpha
    render_visibility
 
-   ray_resampling
    pack_data
    unpack_data
    
\ No newline at end of file

docs/source/examples/ngp.rst

@@ -7,7 +7,7 @@ See code `examples/train_ngp_nerf.py` at our `github repository`_ for details.
 
 Benchmarks
 ------------
-*updated on 2022-10-12*
+*updated on 2023-03-14*
 
 Here we trained a `Instant-NGP Nerf`_ model on the `Nerf-Synthetic dataset`_. We follow the same
 settings with the Instant-NGP paper, which uses train split for training and test split for
@@ -30,11 +30,15 @@ memory footprint is about 3GB.
 +-----------------------+-------+-------+---------+-------+-------+-------+-------+-------+-------+
 |(training time)        | 309s  | 258s  | 256s    | 316s  | 292s  | 207s  | 218s  | 250s  | 263s  |
 +-----------------------+-------+-------+---------+-------+-------+-------+-------+-------+-------+
-|Ours 20k steps         | 35.50 | 36.16 | 29.14   | 35.23 | 37.15 | 31.71 | 24.88 | 29.91 | 32.46 |
+|Ours (occ) 20k steps   | 35.81 | 36.87 | 29.59   | 35.70 | 37.45 | 33.63 | 24.98 | 30.64 | 33.08 |
 +-----------------------+-------+-------+---------+-------+-------+-------+-------+-------+-------+
-|(training time)        | 287s  | 274s  | 269s    | 317s  | 269s  | 244s  | 249s  | 257s  | 271s  |
+|(training time)        | 288s  | 255s  | 247s    | 319s  | 274s  | 238s  | 247s  | 252s  | 265s  |
++-----------------------+-------+-------+---------+-------+-------+-------+-------+-------+-------+
+|Ours (prop) 20k steps  | 34.06 | 34.32 | 27.93   | 34.27 | 36.47 | 31.39 | 24.39 | 30.57 | 31.68 |
++-----------------------+-------+-------+---------+-------+-------+-------+-------+-------+-------+
+|(training time)        | 238s  | 236s  | 250s    | 235s  | 235s  | 236s  | 236s  | 236s  | 240s  |
 +-----------------------+-------+-------+---------+-------+-------+-------+-------+-------+-------+
 
 .. _`Instant-NGP Nerf`: https://github.com/NVlabs/instant-ngp/tree/51e4107edf48338e9ab0316d56a222e0adf87143
-.. _`github repository`: https://github.com/KAIR-BAIR/nerfacc/tree/76c0f9817da4c9c8b5ccf827eb069ee2ce854b75
+.. _`github repository`: https://github.com/KAIR-BAIR/nerfacc/
 .. _`Nerf-Synthetic dataset`: https://drive.google.com/drive/folders/1JDdLGDruGNXWnM1eqY1FNL9PlStjaKWi

docs/source/examples/unbounded.rst

@@ -5,7 +5,7 @@ See code `examples/train_ngp_nerf.py` at our `github repository`_ for details.
 
 Benchmarks
 ------------
-*updated on 2022-11-07*
+*updated on 2023-03-14*
 
 Here we trained a `Instant-NGP Nerf`_  on the `MipNerf360`_ dataset. We used train 
 split for training and test split for evaluation. Our experiments are conducted on a 
@@ -32,12 +32,19 @@ that takes from `MipNerf360`_.
 +----------------------+-------+-------+-------+-------+-------+-------+-------+-------+
 | MipNerf360 (~days)   | 26.98 | 24.37 | 33.46 | 29.55 | 32.23 | 31.63 | 26.40 | 29.23 |
 +----------------------+-------+-------+-------+-------+-------+-------+-------+-------+
-| Ours (~20 mins)      | 25.41 | 22.97 | 30.71 | 27.34 | 30.32 | 31.00 | 23.43 | 27.31 |
+| Ours (occ)           | 24.76 | 22.38 | 29.72 | 26.80 | 28.02 | 30.67 | 22.39 | 26.39 |
 +----------------------+-------+-------+-------+-------+-------+-------+-------+-------+
-| Ours (Training time) | 25min | 17min | 19min | 23min | 28min | 20min | 17min | 21min |
+| Ours (Training time) | 323s  | 302s  | 300s  | 337s  | 347s  | 320s  | 322s  | 322s  |
 +----------------------+-------+-------+-------+-------+-------+-------+-------+-------+
+| Ours (prop)          | 25.44 | 23.21 | 30.62 | 26.75 | 30.63 | 30.93 | 25.20 | 27.54 |
++----------------------+-------+-------+-------+-------+-------+-------+-------+-------+
+| Ours (Training time) | 308s  | 304s  | 308s  | 306s  | 313s  | 301s  | 287s  | 304s  |
++----------------------+-------+-------+-------+-------+-------+-------+-------+-------+
+
+Note `Ours (prop)` is basically a `Nerfacto_` model.
 
 .. _`Instant-NGP Nerf`: https://arxiv.org/abs/2201.05989
 .. _`MipNerf360`: https://arxiv.org/abs/2111.12077
 .. _`Nerf++`: https://arxiv.org/abs/2010.07492
-.. _`github repository`: https://github.com/KAIR-BAIR/nerfacc/tree/76c0f9817da4c9c8b5ccf827eb069ee2ce854b75
+.. _`github repository`: https://github.com/KAIR-BAIR/nerfacc/
+.. _`Nerfacto`: https://github.com/nerfstudio-project/nerfstudio/blob/main/nerfstudio/models/nerfacto.py
\ No newline at end of file

docs/source/index.rst

@@ -123,7 +123,7 @@ Links:
 .. toctree::
    :glob:
    :maxdepth: 1
-   :caption: Example Usages
+   :caption: Example Usages and Benchmarks
 
    examples/*
 

examples/datasets/dnerf_synthetic.py

@@ -86,7 +86,7 @@ class SubjectLoader(torch.utils.data.Dataset):
         near: float = None,
         far: float = None,
         batch_over_images: bool = True,
-        device: str = "cuda:0",
+        device: str = "cpu",
     ):
         super().__init__()
         assert split in self.SPLITS, "%s" % split

examples/datasets/nerf_360_v2.py

@@ -22,7 +22,7 @@ sys.path.insert(
 from scene_manager import SceneManager
 
 
-def _load_colmap(root_fp: str, subject_id: str, split: str, factor: int = 1):
+def _load_colmap(root_fp: str, subject_id: str, factor: int = 1):
     assert factor in [1, 2, 4, 8]
 
     data_dir = os.path.join(root_fp, subject_id)
@@ -134,12 +134,66 @@ def _load_colmap(root_fp: str, subject_id: str, split: str, factor: int = 1):
         "test": all_indices[all_indices % 8 == 0],
         "train": all_indices[all_indices % 8 != 0],
     }
-    indices = split_indices[split]
-    # All per-image quantities must be re-indexed using the split indices.
-    images = images[indices]
-    camtoworlds = camtoworlds[indices]
+    return images, camtoworlds, K, split_indices
+
+
+def similarity_from_cameras(c2w, strict_scaling):
+    """
+    reference: nerf-factory
+    Get a similarity transform to normalize dataset
+    from c2w (OpenCV convention) cameras
+    :param c2w: (N, 4)
+    :return T (4,4) , scale (float)
+    """
+    t = c2w[:, :3, 3]
+    R = c2w[:, :3, :3]
+
+    # (1) Rotate the world so that z+ is the up axis
+    # we estimate the up axis by averaging the camera up axes
+    ups = np.sum(R * np.array([0, -1.0, 0]), axis=-1)
+    world_up = np.mean(ups, axis=0)
+    world_up /= np.linalg.norm(world_up)
+
+    up_camspace = np.array([0.0, -1.0, 0.0])
+    c = (up_camspace * world_up).sum()
+    cross = np.cross(world_up, up_camspace)
+    skew = np.array(
+        [
+            [0.0, -cross[2], cross[1]],
+            [cross[2], 0.0, -cross[0]],
+            [-cross[1], cross[0], 0.0],
+        ]
+    )
+    if c > -1:
+        R_align = np.eye(3) + skew + (skew @ skew) * 1 / (1 + c)
+    else:
+        # In the unlikely case the original data has y+ up axis,
+        # rotate 180-deg about x axis
+        R_align = np.array([[-1.0, 0.0, 0.0], [0.0, 1.0, 0.0], [0.0, 0.0, 1.0]])
+
+    #  R_align = np.eye(3) # DEBUG
+    R = R_align @ R
+    fwds = np.sum(R * np.array([0, 0.0, 1.0]), axis=-1)
+    t = (R_align @ t[..., None])[..., 0]
+
+    # (2) Recenter the scene using camera center rays
+    # find the closest point to the origin for each camera's center ray
+    nearest = t + (fwds * -t).sum(-1)[:, None] * fwds
 
-    return images, camtoworlds, K
+    # median for more robustness
+    translate = -np.median(nearest, axis=0)
+
+    #  translate = -np.mean(t, axis=0)  # DEBUG
+
+    transform = np.eye(4)
+    transform[:3, 3] = translate
+    transform[:3, :3] = R_align
+
+    # (3) Rescale the scene using camera distances
+    scale_fn = np.max if strict_scaling else np.median
+    scale = 1.0 / scale_fn(np.linalg.norm(t + translate, axis=-1))
+
+    return transform, scale
 
 
 class SubjectLoader(torch.utils.data.Dataset):
@@ -169,7 +223,7 @@ class SubjectLoader(torch.utils.data.Dataset):
         far: float = None,
         batch_over_images: bool = True,
         factor: int = 1,
-        device: str = "cuda:0",
+        device: str = "cpu",
     ):
         super().__init__()
         assert split in self.SPLITS, "%s" % split
@@ -184,14 +238,25 @@ class SubjectLoader(torch.utils.data.Dataset):
         )
         self.color_bkgd_aug = color_bkgd_aug
         self.batch_over_images = batch_over_images
-        self.images, self.camtoworlds, self.K = _load_colmap(
-            root_fp, subject_id, split, factor
+        self.images, self.camtoworlds, self.K, split_indices = _load_colmap(
+            root_fp, subject_id, factor
+        )
+        # normalize the scene
+        T, sscale = similarity_from_cameras(
+            self.camtoworlds, strict_scaling=False
         )
-        self.images = torch.from_numpy(self.images).to(device).to(torch.uint8)
+        self.camtoworlds = np.einsum("nij, ki -> nkj", self.camtoworlds, T)
+        self.camtoworlds[:, :3, 3] *= sscale
+        # split
+        indices = split_indices[split]
+        self.images = self.images[indices]
+        self.camtoworlds = self.camtoworlds[indices]
+        # to tensor
+        self.images = torch.from_numpy(self.images).to(torch.uint8).to(device)
         self.camtoworlds = (
-            torch.from_numpy(self.camtoworlds).to(device).to(torch.float32)
+            torch.from_numpy(self.camtoworlds).to(torch.float32).to(device)
         )
-        self.K = torch.tensor(self.K).to(device).to(torch.float32)
+        self.K = torch.tensor(self.K).to(torch.float32).to(device)
         self.height, self.width = self.images.shape[1:3]
 
     def __len__(self):
@@ -275,7 +340,7 @@ class SubjectLoader(torch.utils.data.Dataset):
             value=(-1.0 if self.OPENGL_CAMERA else 1.0),
         )  # [num_rays, 3]
 
-        # [n_cams, height, width, 3]
+        # [num_rays, 3]
         directions = (camera_dirs[:, None, :] * c2w[:, :3, :3]).sum(dim=-1)
         origins = torch.broadcast_to(c2w[:, :3, -1], directions.shape)
         viewdirs = directions / torch.linalg.norm(

examples/datasets/nerf_synthetic.py

@@ -79,7 +79,7 @@ class SubjectLoader(torch.utils.data.Dataset):
         near: float = None,
         far: float = None,
         batch_over_images: bool = True,
-        device: str = "cuda:0",
+        device: torch.device = torch.device("cpu"),
     ):
         super().__init__()
         assert split in self.SPLITS, "%s" % split
@@ -110,10 +110,8 @@ class SubjectLoader(torch.utils.data.Dataset):
             self.images, self.camtoworlds, self.focal = _load_renderings(
                 root_fp, subject_id, split
             )
-        self.images = torch.from_numpy(self.images).to(device).to(torch.uint8)
-        self.camtoworlds = (
-            torch.from_numpy(self.camtoworlds).to(device).to(torch.float32)
-        )
+        self.images = torch.from_numpy(self.images).to(torch.uint8)
+        self.camtoworlds = torch.from_numpy(self.camtoworlds).to(torch.float32)
         self.K = torch.tensor(
             [
                 [self.focal, 0, self.WIDTH / 2.0],
@@ -121,8 +119,10 @@ class SubjectLoader(torch.utils.data.Dataset):
                 [0, 0, 1],
             ],
             dtype=torch.float32,
-            device=device,
         )  # (3, 3)
+        self.images = self.images.to(device)
+        self.camtoworlds = self.camtoworlds.to(device)
+        self.K = self.K.to(device)
         assert self.images.shape[1:3] == (self.HEIGHT, self.WIDTH)
 
     def __len__(self):

examples/radiance_fields/ngp.py

@@ -4,6 +4,7 @@ Copyright (c) 2022 Ruilong Li, UC Berkeley.
 
 from typing import Callable, List, Union
 
+import numpy as np
 import torch
 from torch.autograd import Function
 from torch.cuda.amp import custom_bwd, custom_fwd
@@ -41,13 +42,15 @@ trunc_exp = _TruncExp.apply
 def contract_to_unisphere(
     x: torch.Tensor,
     aabb: torch.Tensor,
+    ord: Union[str, int] = 2,
+    #  ord: Union[float, int] = float("inf"),
     eps: float = 1e-6,
     derivative: bool = False,
 ):
     aabb_min, aabb_max = torch.split(aabb, 3, dim=-1)
     x = (x - aabb_min) / (aabb_max - aabb_min)
     x = x * 2 - 1  # aabb is at [-1, 1]
-    mag = x.norm(dim=-1, keepdim=True)
+    mag = torch.linalg.norm(x, ord=ord, dim=-1, keepdim=True)
     mask = mag.squeeze(-1) > 1
 
     if derivative:
@@ -63,8 +66,8 @@ def contract_to_unisphere(
         return x
 
 
-class NGPradianceField(torch.nn.Module):
-    """Instance-NGP radiance Field"""
+class NGPRadianceField(torch.nn.Module):
+    """Instance-NGP Radiance Field"""
 
     def __init__(
         self,
@@ -73,6 +76,8 @@ class NGPradianceField(torch.nn.Module):
         use_viewdirs: bool = True,
         density_activation: Callable = lambda x: trunc_exp(x - 1),
         unbounded: bool = False,
+        base_resolution: int = 16,
+        max_resolution: int = 4096,
         geo_feat_dim: int = 15,
         n_levels: int = 16,
         log2_hashmap_size: int = 19,
@@ -85,9 +90,15 @@ class NGPradianceField(torch.nn.Module):
         self.use_viewdirs = use_viewdirs
         self.density_activation = density_activation
         self.unbounded = unbounded
-
+        self.base_resolution = base_resolution
+        self.max_resolution = max_resolution
         self.geo_feat_dim = geo_feat_dim
-        per_level_scale = 1.4472692012786865
+        self.n_levels = n_levels
+        self.log2_hashmap_size = log2_hashmap_size
+
+        per_level_scale = np.exp(
+            (np.log(max_resolution) - np.log(base_resolution)) / (n_levels - 1)
+        ).tolist()
 
         if self.use_viewdirs:
             self.direction_encoding = tcnn.Encoding(
@@ -113,7 +124,7 @@ class NGPradianceField(torch.nn.Module):
                 "n_levels": n_levels,
                 "n_features_per_level": 2,
                 "log2_hashmap_size": log2_hashmap_size,
-                "base_resolution": 16,
+                "base_resolution": base_resolution,
                 "per_level_scale": per_level_scale,
             },
             network_config={
@@ -138,7 +149,7 @@ class NGPradianceField(torch.nn.Module):
                 network_config={
                     "otype": "FullyFusedMLP",
                     "activation": "ReLU",
-                    "output_activation": "Sigmoid",
+                    "output_activation": "None",
                     "n_neurons": 64,
                     "n_hidden_layers": 2,
                 },
@@ -168,19 +179,21 @@ class NGPradianceField(torch.nn.Module):
         else:
             return density
 
-    def _query_rgb(self, dir, embedding):
+    def _query_rgb(self, dir, embedding, apply_act: bool = True):
         # tcnn requires directions in the range [0, 1]
         if self.use_viewdirs:
             dir = (dir + 1.0) / 2.0
-            d = self.direction_encoding(dir.view(-1, dir.shape[-1]))
-            h = torch.cat([d, embedding.view(-1, self.geo_feat_dim)], dim=-1)
+            d = self.direction_encoding(dir.reshape(-1, dir.shape[-1]))
+            h = torch.cat([d, embedding.reshape(-1, self.geo_feat_dim)], dim=-1)
         else:
-            h = embedding.view(-1, self.geo_feat_dim)
+            h = embedding.reshape(-1, self.geo_feat_dim)
         rgb = (
             self.mlp_head(h)
-            .view(list(embedding.shape[:-1]) + [3])
+            .reshape(list(embedding.shape[:-1]) + [3])
             .to(embedding)
         )
+        if apply_act:
+            rgb = torch.sigmoid(rgb)
         return rgb
 
     def forward(
@@ -194,4 +207,73 @@ class NGPradianceField(torch.nn.Module):
             ), f"{positions.shape} v.s. {directions.shape}"
             density, embedding = self.query_density(positions, return_feat=True)
             rgb = self._query_rgb(directions, embedding=embedding)
-        return rgb, density
+        return rgb, density  # type: ignore
+
+
+class NGPDensityField(torch.nn.Module):
+    """Instance-NGP Density Field used for resampling"""
+
+    def __init__(
+        self,
+        aabb: Union[torch.Tensor, List[float]],
+        num_dim: int = 3,
+        density_activation: Callable = lambda x: trunc_exp(x - 1),
+        unbounded: bool = False,
+        base_resolution: int = 16,
+        max_resolution: int = 128,
+        n_levels: int = 5,
+        log2_hashmap_size: int = 17,
+    ) -> None:
+        super().__init__()
+        if not isinstance(aabb, torch.Tensor):
+            aabb = torch.tensor(aabb, dtype=torch.float32)
+        self.register_buffer("aabb", aabb)
+        self.num_dim = num_dim
+        self.density_activation = density_activation
+        self.unbounded = unbounded
+        self.base_resolution = base_resolution
+        self.max_resolution = max_resolution
+        self.n_levels = n_levels
+        self.log2_hashmap_size = log2_hashmap_size
+
+        per_level_scale = np.exp(
+            (np.log(max_resolution) - np.log(base_resolution)) / (n_levels - 1)
+        ).tolist()
+
+        self.mlp_base = tcnn.NetworkWithInputEncoding(
+            n_input_dims=num_dim,
+            n_output_dims=1,
+            encoding_config={
+                "otype": "HashGrid",
+                "n_levels": n_levels,
+                "n_features_per_level": 2,
+                "log2_hashmap_size": log2_hashmap_size,
+                "base_resolution": base_resolution,
+                "per_level_scale": per_level_scale,
+            },
+            network_config={
+                "otype": "FullyFusedMLP",
+                "activation": "ReLU",
+                "output_activation": "None",
+                "n_neurons": 64,
+                "n_hidden_layers": 1,
+            },
+        )
+
+    def forward(self, positions: torch.Tensor):
+        if self.unbounded:
+            positions = contract_to_unisphere(positions, self.aabb)
+        else:
+            aabb_min, aabb_max = torch.split(self.aabb, self.num_dim, dim=-1)
+            positions = (positions - aabb_min) / (aabb_max - aabb_min)
+        selector = ((positions > 0.0) & (positions < 1.0)).all(dim=-1)
+        density_before_activation = (
+            self.mlp_base(positions.view(-1, self.num_dim))
+            .view(list(positions.shape[:-1]) + [1])
+            .to(positions)
+        )
+        density = (
+            self.density_activation(density_before_activation)
+            * selector[..., None]
+        )
+        return density

examples/requirements.txt

@@ -4,3 +4,4 @@ imageio
 numpy
 tqdm
 scipy
+lpips
\ No newline at end of file

examples/train_mlp_dnerf.py

@@ -16,28 +16,26 @@ from datasets.dnerf_synthetic import SubjectLoader
 from radiance_fields.mlp import DNeRFRadianceField
 from utils import render_image, set_random_seed
 
-from nerfacc import ContractionType, OccupancyGrid
+from nerfacc import OccupancyGrid
 
-if __name__ == "__main__":
+device = "cuda:0"
+set_random_seed(42)
 
-    device = "cuda:0"
-    set_random_seed(42)
-
-    parser = argparse.ArgumentParser()
-    parser.add_argument(
+parser = argparse.ArgumentParser()
+parser.add_argument(
     "--data_root",
     type=str,
     default=str(pathlib.Path.cwd() / "data/dnerf"),
     help="the root dir of the dataset",
-    )
-    parser.add_argument(
+)
+parser.add_argument(
     "--train_split",
     type=str,
     default="train",
     choices=["train"],
     help="which train split to use",
-    )
-    parser.add_argument(
+)
+parser.add_argument(
     "--scene",
     type=str,
     default="lego",
@@ -53,40 +51,37 @@ if __name__ == "__main__":
         "trex",
     ],
     help="which scene to use",
-    )
-    parser.add_argument(
+)
+parser.add_argument(
     "--aabb",
     type=lambda s: [float(item) for item in s.split(",")],
     default="-1.5,-1.5,-1.5,1.5,1.5,1.5",
     help="delimited list input",
-    )
-    parser.add_argument(
+)
+parser.add_argument(
     "--test_chunk_size",
     type=int,
     default=8192,
-    )
-    parser.add_argument("--cone_angle", type=float, default=0.0)
-    args = parser.parse_args()
-
-    render_n_samples = 1024
-
-    # setup the scene bounding box.
-    contraction_type = ContractionType.AABB
-    scene_aabb = torch.tensor(args.aabb, dtype=torch.float32, device=device)
-    near_plane = None
-    far_plane = None
-    render_step_size = (
-        (scene_aabb[3:] - scene_aabb[:3]).max()
-        * math.sqrt(3)
-        / render_n_samples
-    ).item()
-
-    # setup the radiance field we want to train.
-    max_steps = 30000
-    grad_scaler = torch.cuda.amp.GradScaler(1)
-    radiance_field = DNeRFRadianceField().to(device)
-    optimizer = torch.optim.Adam(radiance_field.parameters(), lr=5e-4)
-    scheduler = torch.optim.lr_scheduler.MultiStepLR(
+)
+parser.add_argument("--cone_angle", type=float, default=0.0)
+args = parser.parse_args()
+
+render_n_samples = 1024
+
+# setup the scene bounding box.
+scene_aabb = torch.tensor(args.aabb, dtype=torch.float32, device=device)
+near_plane = None
+far_plane = None
+render_step_size = (
+    (scene_aabb[3:] - scene_aabb[:3]).max() * math.sqrt(3) / render_n_samples
+).item()
+
+# setup the radiance field we want to train.
+max_steps = 30000
+grad_scaler = torch.cuda.amp.GradScaler(1)
+radiance_field = DNeRFRadianceField().to(device)
+optimizer = torch.optim.Adam(radiance_field.parameters(), lr=5e-4)
+scheduler = torch.optim.lr_scheduler.MultiStepLR(
     optimizer,
     milestones=[
         max_steps // 2,
@@ -95,35 +90,35 @@ if __name__ == "__main__":
         max_steps * 9 // 10,
     ],
     gamma=0.33,
-    )
-    # setup the dataset
-    target_sample_batch_size = 1 << 16
-    grid_resolution = 128
+)
+# setup the dataset
+target_sample_batch_size = 1 << 16
+grid_resolution = 128
 
-    train_dataset = SubjectLoader(
+train_dataset = SubjectLoader(
     subject_id=args.scene,
     root_fp=args.data_root,
     split=args.train_split,
     num_rays=target_sample_batch_size // render_n_samples,
-    )
+    device=device,
+)
 
-    test_dataset = SubjectLoader(
+test_dataset = SubjectLoader(
     subject_id=args.scene,
     root_fp=args.data_root,
     split="test",
     num_rays=None,
-    )
+    device=device,
+)
 
-    occupancy_grid = OccupancyGrid(
-        roi_aabb=args.aabb,
-        resolution=grid_resolution,
-        contraction_type=contraction_type,
-    ).to(device)
+occupancy_grid = OccupancyGrid(
+    roi_aabb=args.aabb, resolution=grid_resolution
+).to(device)
 
-    # training
-    step = 0
-    tic = time.time()
-    for epoch in range(10000000):
+# training
+step = 0
+tic = time.time()
+for epoch in range(10000000):
     for i in range(len(train_dataset)):
         radiance_field.train()
         data = train_dataset[i]
@@ -163,8 +158,7 @@ if __name__ == "__main__":
         # dynamic batch size for rays to keep sample batch size constant.
         num_rays = len(pixels)
         num_rays = int(
-                num_rays
-                * (target_sample_batch_size / float(n_rendering_samples))
+            num_rays * (target_sample_batch_size / float(n_rendering_samples))
         )
         train_dataset.update_num_rays(num_rays)
         alive_ray_mask = acc.squeeze(-1) > 0

examples/train_mlp_nerf.py

@@ -17,26 +17,24 @@ from utils import render_image, set_random_seed
 
 from nerfacc import ContractionType, OccupancyGrid
 
-if __name__ == "__main__":
+device = "cuda:0"
+set_random_seed(42)
 
-    device = "cuda:0"
-    set_random_seed(42)
-
-    parser = argparse.ArgumentParser()
-    parser.add_argument(
+parser = argparse.ArgumentParser()
+parser.add_argument(
     "--data_root",
     type=str,
     default=str(pathlib.Path.cwd() / "data/nerf_synthetic"),
     help="the root dir of the dataset",
-    )
-    parser.add_argument(
+)
+parser.add_argument(
     "--train_split",
     type=str,
     default="trainval",
     choices=["train", "trainval"],
     help="which train split to use",
-    )
-    parser.add_argument(
+)
+parser.add_argument(
     "--scene",
     type=str,
     default="lego",
@@ -54,30 +52,30 @@ if __name__ == "__main__":
         "garden",
     ],
     help="which scene to use",
-    )
-    parser.add_argument(
+)
+parser.add_argument(
     "--aabb",
     type=lambda s: [float(item) for item in s.split(",")],
     default="-1.5,-1.5,-1.5,1.5,1.5,1.5",
     help="delimited list input",
-    )
-    parser.add_argument(
+)
+parser.add_argument(
     "--test_chunk_size",
     type=int,
     default=8192,
-    )
-    parser.add_argument(
+)
+parser.add_argument(
     "--unbounded",
     action="store_true",
     help="whether to use unbounded rendering",
-    )
-    parser.add_argument("--cone_angle", type=float, default=0.0)
-    args = parser.parse_args()
+)
+parser.add_argument("--cone_angle", type=float, default=0.0)
+args = parser.parse_args()
 
-    render_n_samples = 1024
+render_n_samples = 1024
 
-    # setup the scene bounding box.
-    if args.unbounded:
+# setup the scene bounding box.
+if args.unbounded:
     print("Using unbounded rendering")
     contraction_type = ContractionType.UN_BOUNDED_SPHERE
     # contraction_type = ContractionType.UN_BOUNDED_TANH
@@ -85,7 +83,7 @@ if __name__ == "__main__":
     near_plane = 0.2
     far_plane = 1e4
     render_step_size = 1e-2
-    else:
+else:
     contraction_type = ContractionType.AABB
     scene_aabb = torch.tensor(args.aabb, dtype=torch.float32, device=device)
     near_plane = None
@@ -96,12 +94,12 @@ if __name__ == "__main__":
         / render_n_samples
     ).item()
 
-    # setup the radiance field we want to train.
-    max_steps = 50000
-    grad_scaler = torch.cuda.amp.GradScaler(1)
-    radiance_field = VanillaNeRFRadianceField().to(device)
-    optimizer = torch.optim.Adam(radiance_field.parameters(), lr=5e-4)
-    scheduler = torch.optim.lr_scheduler.MultiStepLR(
+# setup the radiance field we want to train.
+max_steps = 50000
+grad_scaler = torch.cuda.amp.GradScaler(1)
+radiance_field = VanillaNeRFRadianceField().to(device)
+optimizer = torch.optim.Adam(radiance_field.parameters(), lr=5e-4)
+scheduler = torch.optim.lr_scheduler.MultiStepLR(
     optimizer,
     milestones=[
         max_steps // 2,
@@ -110,50 +108,50 @@ if __name__ == "__main__":
         max_steps * 9 // 10,
     ],
     gamma=0.33,
-    )
+)
 
-    # setup the dataset
-    train_dataset_kwargs = {}
-    test_dataset_kwargs = {}
-    if args.scene == "garden":
+# setup the dataset
+train_dataset_kwargs = {}
+test_dataset_kwargs = {}
+if args.scene == "garden":
     from datasets.nerf_360_v2 import SubjectLoader
 
     target_sample_batch_size = 1 << 16
     train_dataset_kwargs = {"color_bkgd_aug": "random", "factor": 4}
     test_dataset_kwargs = {"factor": 4}
     grid_resolution = 128
-    else:
+else:
     from datasets.nerf_synthetic import SubjectLoader
 
     target_sample_batch_size = 1 << 16
     grid_resolution = 128
 
-    train_dataset = SubjectLoader(
+train_dataset = SubjectLoader(
     subject_id=args.scene,
     root_fp=args.data_root,
     split=args.train_split,
     num_rays=target_sample_batch_size // render_n_samples,
     **train_dataset_kwargs,
-    )
+)
 
-    test_dataset = SubjectLoader(
+test_dataset = SubjectLoader(
     subject_id=args.scene,
     root_fp=args.data_root,
     split="test",
     num_rays=None,
     **test_dataset_kwargs,
-    )
+)
 
-    occupancy_grid = OccupancyGrid(
+occupancy_grid = OccupancyGrid(
     roi_aabb=args.aabb,
     resolution=grid_resolution,
     contraction_type=contraction_type,
-    ).to(device)
+).to(device)
 
-    # training
-    step = 0
-    tic = time.time()
-    for epoch in range(10000000):
+# training
+step = 0
+tic = time.time()
+for epoch in range(10000000):
     for i in range(len(train_dataset)):
         radiance_field.train()
         data = train_dataset[i]
@@ -189,8 +187,7 @@ if __name__ == "__main__":
         # dynamic batch size for rays to keep sample batch size constant.
         num_rays = len(pixels)
         num_rays = int(
-                num_rays
-                * (target_sample_batch_size / float(n_rendering_samples))
+            num_rays * (target_sample_batch_size / float(n_rendering_samples))
         )
         train_dataset.update_num_rays(num_rays)
         alive_ray_mask = acc.squeeze(-1) > 0

examples/train_ngp_nerf.py

@@ -12,172 +12,155 @@ import numpy as np
 import torch
 import torch.nn.functional as F
 import tqdm
-from radiance_fields.ngp import NGPradianceField
-from utils import render_image, set_random_seed
+from lpips import LPIPS
+from radiance_fields.ngp import NGPRadianceField
+from utils import (
+    MIPNERF360_UNBOUNDED_SCENES,
+    NERF_SYNTHETIC_SCENES,
+    enlarge_aabb,
+    render_image,
+    set_random_seed,
+)
 
-from nerfacc import ContractionType, OccupancyGrid
+from nerfacc import OccupancyGrid
 
-if __name__ == "__main__":
-
-    device = "cuda:0"
-    set_random_seed(42)
-
-    parser = argparse.ArgumentParser()
-    parser.add_argument(
+parser = argparse.ArgumentParser()
+parser.add_argument(
     "--data_root",
     type=str,
-        default=str(pathlib.Path.cwd() / "data"),
+    # default=str(pathlib.Path.cwd() / "data/360_v2"),
+    default=str(pathlib.Path.cwd() / "data/nerf_synthetic"),
     help="the root dir of the dataset",
-    )
-    parser.add_argument(
+)
+parser.add_argument(
     "--train_split",
     type=str,
-        default="trainval",
+    default="train",
     choices=["train", "trainval"],
     help="which train split to use",
-    )
-    parser.add_argument(
+)
+parser.add_argument(
     "--scene",
     type=str,
     default="lego",
-        choices=[
-            # nerf synthetic
-            "chair",
-            "drums",
-            "ficus",
-            "hotdog",
-            "lego",
-            "materials",
-            "mic",
-            "ship",
-            # mipnerf360 unbounded
-            "garden",
-            "bicycle",
-            "bonsai",
-            "counter",
-            "kitchen",
-            "room",
-            "stump",
-        ],
+    choices=NERF_SYNTHETIC_SCENES + MIPNERF360_UNBOUNDED_SCENES,
     help="which scene to use",
-    )
-    parser.add_argument(
-        "--aabb",
-        type=lambda s: [float(item) for item in s.split(",")],
-        default="-1.5,-1.5,-1.5,1.5,1.5,1.5",
-        help="delimited list input",
-    )
-    parser.add_argument(
+)
+parser.add_argument(
     "--test_chunk_size",
     type=int,
     default=8192,
-    )
-    parser.add_argument(
-        "--unbounded",
-        action="store_true",
-        help="whether to use unbounded rendering",
-    )
-    parser.add_argument(
-        "--auto_aabb",
-        action="store_true",
-        help="whether to automatically compute the aabb",
-    )
-    parser.add_argument("--cone_angle", type=float, default=0.0)
-    args = parser.parse_args()
+)
+args = parser.parse_args()
 
-    render_n_samples = 1024
+device = "cuda:0"
+set_random_seed(42)
 
-    # setup the dataset
-    train_dataset_kwargs = {}
-    test_dataset_kwargs = {}
-    if args.unbounded:
+if args.scene in MIPNERF360_UNBOUNDED_SCENES:
     from datasets.nerf_360_v2 import SubjectLoader
 
-        target_sample_batch_size = 1 << 20
+    # training parameters
+    max_steps = 100000
+    init_batch_size = 1024
+    target_sample_batch_size = 1 << 18
+    weight_decay = 0.0
+    # scene parameters
+    aabb = torch.tensor([-1.0, -1.0, -1.0, 1.0, 1.0, 1.0], device=device)
+    near_plane = 0.02
+    far_plane = None
+    # dataset parameters
     train_dataset_kwargs = {"color_bkgd_aug": "random", "factor": 4}
     test_dataset_kwargs = {"factor": 4}
-        grid_resolution = 256
-    else:
+    # model parameters
+    grid_resolution = 128
+    grid_nlvl = 4
+    # render parameters
+    render_step_size = 1e-3
+    alpha_thre = 1e-2
+    cone_angle = 0.004
+
+else:
     from datasets.nerf_synthetic import SubjectLoader
 
+    # training parameters
+    max_steps = 20000
+    init_batch_size = 1024
     target_sample_batch_size = 1 << 18
+    weight_decay = (
+        1e-5 if args.scene in ["materials", "ficus", "drums"] else 1e-6
+    )
+    # scene parameters
+    aabb = torch.tensor([-1.5, -1.5, -1.5, 1.5, 1.5, 1.5], device=device)
+    near_plane = None
+    far_plane = None
+    # dataset parameters
+    train_dataset_kwargs = {}
+    test_dataset_kwargs = {}
+    # model parameters
     grid_resolution = 128
+    grid_nlvl = 1
+    # render parameters
+    render_step_size = 5e-3
+    alpha_thre = 0.0
+    cone_angle = 0.0
 
-    train_dataset = SubjectLoader(
+train_dataset = SubjectLoader(
     subject_id=args.scene,
     root_fp=args.data_root,
     split=args.train_split,
-        num_rays=target_sample_batch_size // render_n_samples,
+    num_rays=init_batch_size,
+    device=device,
     **train_dataset_kwargs,
-    )
+)
 
-    test_dataset = SubjectLoader(
+test_dataset = SubjectLoader(
     subject_id=args.scene,
     root_fp=args.data_root,
     split="test",
     num_rays=None,
+    device=device,
     **test_dataset_kwargs,
-    )
+)
 
-    if args.auto_aabb:
-        camera_locs = torch.cat(
-            [train_dataset.camtoworlds, test_dataset.camtoworlds]
-        )[:, :3, -1]
-        args.aabb = torch.cat(
-            [camera_locs.min(dim=0).values, camera_locs.max(dim=0).values]
-        ).tolist()
-        print("Using auto aabb", args.aabb)
-
-    # setup the scene bounding box.
-    if args.unbounded:
-        print("Using unbounded rendering")
-        contraction_type = ContractionType.UN_BOUNDED_SPHERE
-        # contraction_type = ContractionType.UN_BOUNDED_TANH
-        scene_aabb = None
-        near_plane = 0.2
-        far_plane = 1e4
-        render_step_size = 1e-2
-        alpha_thre = 1e-2
-    else:
-        contraction_type = ContractionType.AABB
-        scene_aabb = torch.tensor(args.aabb, dtype=torch.float32, device=device)
-        near_plane = None
-        far_plane = None
-        render_step_size = (
-            (scene_aabb[3:] - scene_aabb[:3]).max()
-            * math.sqrt(3)
-            / render_n_samples
-        ).item()
-        alpha_thre = 0.0
+# setup scene aabb
+scene_aabb = enlarge_aabb(aabb, 1 << (grid_nlvl - 1))
 
-    # setup the radiance field we want to train.
-    max_steps = 20000
-    grad_scaler = torch.cuda.amp.GradScaler(2**10)
-    radiance_field = NGPradianceField(
-        aabb=args.aabb,
-        unbounded=args.unbounded,
-    ).to(device)
-    optimizer = torch.optim.Adam(
-        radiance_field.parameters(), lr=1e-2, eps=1e-15
-    )
-    scheduler = torch.optim.lr_scheduler.MultiStepLR(
+# setup the radiance field we want to train.
+grad_scaler = torch.cuda.amp.GradScaler(2**10)
+radiance_field = NGPRadianceField(aabb=scene_aabb).to(device)
+optimizer = torch.optim.Adam(
+    radiance_field.parameters(), lr=1e-2, eps=1e-15, weight_decay=weight_decay
+)
+scheduler = torch.optim.lr_scheduler.ChainedScheduler(
+    [
+        torch.optim.lr_scheduler.LinearLR(
+            optimizer, start_factor=0.01, total_iters=100
+        ),
+        torch.optim.lr_scheduler.MultiStepLR(
             optimizer,
-        milestones=[max_steps // 2, max_steps * 3 // 4, max_steps * 9 // 10],
+            milestones=[
+                max_steps // 2,
+                max_steps * 3 // 4,
+                max_steps * 9 // 10,
+            ],
             gamma=0.33,
-    )
+        ),
+    ]
+)
+occupancy_grid = OccupancyGrid(
+    roi_aabb=aabb, resolution=grid_resolution, levels=grid_nlvl
+).to(device)
 
-    occupancy_grid = OccupancyGrid(
-        roi_aabb=args.aabb,
-        resolution=grid_resolution,
-        contraction_type=contraction_type,
-    ).to(device)
+lpips_net = LPIPS(net="vgg").to(device)
+lpips_norm_fn = lambda x: x[None, ...].permute(0, 3, 1, 2) * 2 - 1
+lpips_fn = lambda x, y: lpips_net(lpips_norm_fn(x), lpips_norm_fn(y)).mean()
 
-    # training
-    step = 0
-    tic = time.time()
-    for epoch in range(10000000):
-        for i in range(len(train_dataset)):
+# training
+tic = time.time()
+for step in range(max_steps + 1):
     radiance_field.train()
+
+    i = torch.randint(0, len(train_dataset), (1,)).item()
     data = train_dataset[i]
 
     render_bkgd = data["color_bkgd"]
@@ -185,61 +168,42 @@ if __name__ == "__main__":
     pixels = data["pixels"]
 
     def occ_eval_fn(x):
-                if args.cone_angle > 0.0:
-                    # randomly sample a camera for computing step size.
-                    camera_ids = torch.randint(
-                        0, len(train_dataset), (x.shape[0],), device=device
-                    )
-                    origins = train_dataset.camtoworlds[camera_ids, :3, -1]
-                    t = (origins - x).norm(dim=-1, keepdim=True)
-                    # compute actual step size used in marching, based on the distance to the camera.
-                    step_size = torch.clamp(
-                        t * args.cone_angle, min=render_step_size
-                    )
-                    # filter out the points that are not in the near far plane.
-                    if (near_plane is not None) and (far_plane is not None):
-                        step_size = torch.where(
-                            (t > near_plane) & (t < far_plane),
-                            step_size,
-                            torch.zeros_like(step_size),
-                        )
-                else:
-                    step_size = render_step_size
-                # compute occupancy
         density = radiance_field.query_density(x)
-                return density * step_size
+        return density * render_step_size
 
     # update occupancy grid
-            occupancy_grid.every_n_step(step=step, occ_eval_fn=occ_eval_fn)
+    occupancy_grid.every_n_step(
+        step=step,
+        occ_eval_fn=occ_eval_fn,
+        occ_thre=1e-2,
+    )
 
     # render
     rgb, acc, depth, n_rendering_samples = render_image(
         radiance_field,
         occupancy_grid,
         rays,
-                scene_aabb,
+        scene_aabb=scene_aabb,
         # rendering options
         near_plane=near_plane,
-                far_plane=far_plane,
         render_step_size=render_step_size,
         render_bkgd=render_bkgd,
-                cone_angle=args.cone_angle,
+        cone_angle=cone_angle,
         alpha_thre=alpha_thre,
     )
     if n_rendering_samples == 0:
         continue
 
+    if target_sample_batch_size > 0:
         # dynamic batch size for rays to keep sample batch size constant.
         num_rays = len(pixels)
         num_rays = int(
-                num_rays
-                * (target_sample_batch_size / float(n_rendering_samples))
+            num_rays * (target_sample_batch_size / float(n_rendering_samples))
         )
         train_dataset.update_num_rays(num_rays)
-            alive_ray_mask = acc.squeeze(-1) > 0
 
     # compute loss
-            loss = F.smooth_l1_loss(rgb[alive_ray_mask], pixels[alive_ray_mask])
+    loss = F.smooth_l1_loss(rgb, pixels)
 
     optimizer.zero_grad()
     # do not unscale it because we are using Adam.
@@ -247,14 +211,15 @@ if __name__ == "__main__":
     optimizer.step()
     scheduler.step()
 
-            if step % 10000 == 0:
+    if step % 5000 == 0:
         elapsed_time = time.time() - tic
-                loss = F.mse_loss(rgb[alive_ray_mask], pixels[alive_ray_mask])
+        loss = F.mse_loss(rgb, pixels)
+        psnr = -10.0 * torch.log(loss) / np.log(10.0)
         print(
             f"elapsed_time={elapsed_time:.2f}s | step={step} | "
-                    f"loss={loss:.5f} | "
-                    f"alive_ray_mask={alive_ray_mask.long().sum():d} | "
-                    f"n_rendering_samples={n_rendering_samples:d} | num_rays={len(pixels):d} |"
+            f"loss={loss:.5f} | psnr={psnr:.2f} | "
+            f"n_rendering_samples={n_rendering_samples:d} | num_rays={len(pixels):d} | "
+            f"max_depth={depth.max():.3f} | "
         )
 
     if step > 0 and step % max_steps == 0:
@@ -262,6 +227,7 @@ if __name__ == "__main__":
         radiance_field.eval()
 
         psnrs = []
+        lpips = []
         with torch.no_grad():
             for i in tqdm.tqdm(range(len(test_dataset))):
                 data = test_dataset[i]
@@ -274,13 +240,12 @@ if __name__ == "__main__":
                     radiance_field,
                     occupancy_grid,
                     rays,
-                            scene_aabb,
+                    scene_aabb=scene_aabb,
                     # rendering options
                     near_plane=near_plane,
-                            far_plane=far_plane,
                     render_step_size=render_step_size,
                     render_bkgd=render_bkgd,
-                            cone_angle=args.cone_angle,
+                    cone_angle=cone_angle,
                     alpha_thre=alpha_thre,
                     # test options
                     test_chunk_size=args.test_chunk_size,
@@ -288,21 +253,18 @@ if __name__ == "__main__":
                 mse = F.mse_loss(rgb, pixels)
                 psnr = -10.0 * torch.log(mse) / np.log(10.0)
                 psnrs.append(psnr.item())
-                        # imageio.imwrite(
-                        #     "acc_binary_test.png",
-                        #     ((acc > 0).float().cpu().numpy() * 255).astype(np.uint8),
-                        # )
+                lpips.append(lpips_fn(rgb, pixels).item())
+                # if i == 0:
                 #     imageio.imwrite(
                 #         "rgb_test.png",
                 #         (rgb.cpu().numpy() * 255).astype(np.uint8),
                 #     )
-                        # break
+                #     imageio.imwrite(
+                #         "rgb_error.png",
+                #         (
+                #             (rgb - pixels).norm(dim=-1).cpu().numpy() * 255
+                #         ).astype(np.uint8),
+                #     )
         psnr_avg = sum(psnrs) / len(psnrs)
-                print(f"evaluation: psnr_avg={psnr_avg}")
-                train_dataset.training = True
-
-            if step == max_steps:
-                print("training stops")
-                exit()
-
-            step += 1
+        lpips_avg = sum(lpips) / len(lpips)
+        print(f"evaluation: psnr_avg={psnr_avg}, lpips_avg={lpips_avg}")

examples/train_ngp_nerf_prop.py

+"""
+Copyright (c) 2022 Ruilong Li, UC Berkeley.
+"""
+
+import argparse
+import itertools
+import pathlib
+import time
+
+import imageio
+import numpy as np
+import torch
+import torch.nn.functional as F
+import tqdm
+from lpips import LPIPS
+from radiance_fields.ngp import NGPDensityField, NGPRadianceField
+from utils import (
+    MIPNERF360_UNBOUNDED_SCENES,
+    NERF_SYNTHETIC_SCENES,
+    render_image_proposal,
+    set_random_seed,
+)
+
+from nerfacc.proposal import (
+    compute_prop_loss,
+    get_proposal_annealing_fn,
+    get_proposal_requires_grad_fn,
+)
+
+parser = argparse.ArgumentParser()
+parser.add_argument(
+    "--data_root",
+    type=str,
+    # default=str(pathlib.Path.cwd() / "data/360_v2"),
+    default=str(pathlib.Path.cwd() / "data/nerf_synthetic"),
+    help="the root dir of the dataset",
+)
+parser.add_argument(
+    "--train_split",
+    type=str,
+    default="train",
+    choices=["train", "trainval"],
+    help="which train split to use",
+)
+parser.add_argument(
+    "--scene",
+    type=str,
+    default="lego",
+    choices=NERF_SYNTHETIC_SCENES + MIPNERF360_UNBOUNDED_SCENES,
+    help="which scene to use",
+)
+parser.add_argument(
+    "--test_chunk_size",
+    type=int,
+    default=8192,
+)
+args = parser.parse_args()
+
+device = "cuda:0"
+set_random_seed(42)
+
+if args.scene in MIPNERF360_UNBOUNDED_SCENES:
+    from datasets.nerf_360_v2 import SubjectLoader
+
+    # training parameters
+    max_steps = 100000
+    init_batch_size = 4096
+    weight_decay = 0.0
+    # scene parameters
+    unbounded = True
+    aabb = torch.tensor([-1.0, -1.0, -1.0, 1.0, 1.0, 1.0], device=device)
+    near_plane = 0.2  # TODO: Try 0.02
+    far_plane = 1e3
+    # dataset parameters
+    train_dataset_kwargs = {"color_bkgd_aug": "random", "factor": 4}
+    test_dataset_kwargs = {"factor": 4}
+    # model parameters
+    proposal_networks = [
+        NGPDensityField(
+            aabb=aabb,
+            unbounded=unbounded,
+            n_levels=5,
+            max_resolution=128,
+        ).to(device),
+        NGPDensityField(
+            aabb=aabb,
+            unbounded=unbounded,
+            n_levels=5,
+            max_resolution=256,
+        ).to(device),
+    ]
+    # render parameters
+    num_samples = 48
+    num_samples_per_prop = [256, 96]
+    sampling_type = "lindisp"
+    opaque_bkgd = True
+
+else:
+    from datasets.nerf_synthetic import SubjectLoader
+
+    # training parameters
+    max_steps = 20000
+    init_batch_size = 4096
+    weight_decay = (
+        1e-5 if args.scene in ["materials", "ficus", "drums"] else 1e-6
+    )
+    # scene parameters
+    unbounded = False
+    aabb = torch.tensor([-1.5, -1.5, -1.5, 1.5, 1.5, 1.5], device=device)
+    near_plane = 2.0
+    far_plane = 6.0
+    # dataset parameters
+    train_dataset_kwargs = {}
+    test_dataset_kwargs = {}
+    # model parameters
+    proposal_networks = [
+        NGPDensityField(
+            aabb=aabb,
+            unbounded=unbounded,
+            n_levels=5,
+            max_resolution=128,
+        ).to(device),
+    ]
+    # render parameters
+    num_samples = 64
+    num_samples_per_prop = [128]
+    sampling_type = "uniform"
+    opaque_bkgd = False
+
+train_dataset = SubjectLoader(
+    subject_id=args.scene,
+    root_fp=args.data_root,
+    split=args.train_split,
+    num_rays=init_batch_size,
+    device=device,
+    **train_dataset_kwargs,
+)
+
+test_dataset = SubjectLoader(
+    subject_id=args.scene,
+    root_fp=args.data_root,
+    split="test",
+    num_rays=None,
+    device=device,
+    **test_dataset_kwargs,
+)
+
+# setup the radiance field we want to train.
+grad_scaler = torch.cuda.amp.GradScaler(2**10)
+radiance_field = NGPRadianceField(aabb=aabb, unbounded=unbounded).to(device)
+optimizer = torch.optim.Adam(
+    itertools.chain(
+        radiance_field.parameters(),
+        *[p.parameters() for p in proposal_networks],
+    ),
+    lr=1e-2,
+    eps=1e-15,
+    weight_decay=weight_decay,
+)
+scheduler = torch.optim.lr_scheduler.ChainedScheduler(
+    [
+        torch.optim.lr_scheduler.LinearLR(
+            optimizer, start_factor=0.01, total_iters=100
+        ),
+        torch.optim.lr_scheduler.MultiStepLR(
+            optimizer,
+            milestones=[
+                max_steps // 2,
+                max_steps * 3 // 4,
+                max_steps * 9 // 10,
+            ],
+            gamma=0.33,
+        ),
+    ]
+)
+proposal_requires_grad_fn = get_proposal_requires_grad_fn()
+proposal_annealing_fn = get_proposal_annealing_fn()
+
+lpips_net = LPIPS(net="vgg").to(device)
+lpips_norm_fn = lambda x: x[None, ...].permute(0, 3, 1, 2) * 2 - 1
+lpips_fn = lambda x, y: lpips_net(lpips_norm_fn(x), lpips_norm_fn(y)).mean()
+
+# training
+tic = time.time()
+for step in range(max_steps + 1):
+    radiance_field.train()
+    for p in proposal_networks:
+        p.train()
+
+    i = torch.randint(0, len(train_dataset), (1,)).item()
+    data = train_dataset[i]
+
+    render_bkgd = data["color_bkgd"]
+    rays = data["rays"]
+    pixels = data["pixels"]
+
+    # render
+    (
+        rgb,
+        acc,
+        depth,
+        weights_per_level,
+        s_vals_per_level,
+    ) = render_image_proposal(
+        radiance_field,
+        proposal_networks,
+        rays,
+        scene_aabb=None,
+        # rendering options
+        num_samples=num_samples,
+        num_samples_per_prop=num_samples_per_prop,
+        near_plane=near_plane,
+        far_plane=far_plane,
+        sampling_type=sampling_type,
+        opaque_bkgd=opaque_bkgd,
+        render_bkgd=render_bkgd,
+        # train options
+        proposal_requires_grad=proposal_requires_grad_fn(step),
+        proposal_annealing=proposal_annealing_fn(step),
+    )
+
+    # compute loss
+    loss = F.smooth_l1_loss(rgb, pixels)
+    loss_prop = compute_prop_loss(s_vals_per_level, weights_per_level)
+    loss = loss + loss_prop
+
+    optimizer.zero_grad()
+    # do not unscale it because we are using Adam.
+    grad_scaler.scale(loss).backward()
+    optimizer.step()
+    scheduler.step()
+
+    if step % 5000 == 0:
+        elapsed_time = time.time() - tic
+        loss = F.mse_loss(rgb, pixels)
+        psnr = -10.0 * torch.log(loss) / np.log(10.0)
+        print(
+            f"elapsed_time={elapsed_time:.2f}s | step={step} | "
+            f"loss={loss:.5f} | psnr={psnr:.2f} | "
+            f"num_rays={len(pixels):d} | "
+            f"max_depth={depth.max():.3f} | "
+        )
+
+    if step > 0 and step % max_steps == 0:
+        # evaluation
+        radiance_field.eval()
+        for p in proposal_networks:
+            p.eval()
+
+        psnrs = []
+        lpips = []
+        with torch.no_grad():
+            for i in tqdm.tqdm(range(len(test_dataset))):
+                data = test_dataset[i]
+                render_bkgd = data["color_bkgd"]
+                rays = data["rays"]
+                pixels = data["pixels"]
+
+                # rendering
+                rgb, acc, depth, _, _, = render_image_proposal(
+                    radiance_field,
+                    proposal_networks,
+                    rays,
+                    scene_aabb=None,
+                    # rendering options
+                    num_samples=num_samples,
+                    num_samples_per_prop=num_samples_per_prop,
+                    near_plane=near_plane,
+                    far_plane=far_plane,
+                    sampling_type=sampling_type,
+                    opaque_bkgd=opaque_bkgd,
+                    render_bkgd=render_bkgd,
+                    proposal_annealing=proposal_annealing_fn(step),
+                    # test options
+                    test_chunk_size=args.test_chunk_size,
+                )
+                mse = F.mse_loss(rgb, pixels)
+                psnr = -10.0 * torch.log(mse) / np.log(10.0)
+                psnrs.append(psnr.item())
+                lpips.append(lpips_fn(rgb, pixels).item())
+                # if i == 0:
+                #     imageio.imwrite(
+                #         "rgb_test.png",
+                #         (rgb.cpu().numpy() * 255).astype(np.uint8),
+                #     )
+                #     imageio.imwrite(
+                #         "rgb_error.png",
+                #         (
+                #             (rgb - pixels).norm(dim=-1).cpu().numpy() * 255
+                #         ).astype(np.uint8),
+                #     )
+        psnr_avg = sum(psnrs) / len(psnrs)
+        lpips_avg = sum(lpips) / len(lpips)
+        print(f"evaluation: psnr_avg={psnr_avg}, lpips_avg={lpips_avg}")

examples/utils.py

@@ -3,13 +3,35 @@ Copyright (c) 2022 Ruilong Li, UC Berkeley.
 """
 
 import random
-from typing import Optional
+from typing import Literal, Optional, Sequence
 
 import numpy as np
 import torch
 from datasets.utils import Rays, namedtuple_map
+from torch.utils.data._utils.collate import collate, default_collate_fn_map
 
 from nerfacc import OccupancyGrid, ray_marching, rendering
+from nerfacc.proposal import rendering as rendering_proposal
+
+NERF_SYNTHETIC_SCENES = [
+    "chair",
+    "drums",
+    "ficus",
+    "hotdog",
+    "lego",
+    "materials",
+    "mic",
+    "ship",
+]
+MIPNERF360_UNBOUNDED_SCENES = [
+    "garden",
+    "bicycle",
+    "bonsai",
+    "counter",
+    "kitchen",
+    "room",
+    "stump",
+]
 
 
 def set_random_seed(seed):
@@ -18,6 +40,12 @@ def set_random_seed(seed):
     torch.manual_seed(seed)
 
 
+def enlarge_aabb(aabb, factor: float) -> torch.Tensor:
+    center = (aabb[:3] + aabb[3:]) / 2
+    extent = (aabb[3:] - aabb[:3]) / 2
+    return torch.cat([center - extent * factor, center + extent * factor])
+
+
 def render_image(
     # scene
     radiance_field: torch.nn.Module,
@@ -116,3 +144,99 @@ def render_image(
         depths.view((*rays_shape[:-1], -1)),
         sum(n_rendering_samples),
     )
+
+
+def render_image_proposal(
+    # scene
+    radiance_field: torch.nn.Module,
+    proposal_networks: Sequence[torch.nn.Module],
+    rays: Rays,
+    scene_aabb: torch.Tensor,
+    # rendering options
+    num_samples: int,
+    num_samples_per_prop: Sequence[int],
+    near_plane: Optional[float] = None,
+    far_plane: Optional[float] = None,
+    sampling_type: Literal["uniform", "lindisp"] = "lindisp",
+    opaque_bkgd: bool = True,
+    render_bkgd: Optional[torch.Tensor] = None,
+    # train options
+    proposal_requires_grad: bool = False,
+    proposal_annealing: float = 1.0,
+    # test options
+    test_chunk_size: int = 8192,
+):
+    """Render the pixels of an image."""
+    rays_shape = rays.origins.shape
+    if len(rays_shape) == 3:
+        height, width, _ = rays_shape
+        num_rays = height * width
+        rays = namedtuple_map(
+            lambda r: r.reshape([num_rays] + list(r.shape[2:])), rays
+        )
+    else:
+        num_rays, _ = rays_shape
+
+    def prop_sigma_fn(t_starts, t_ends, proposal_network):
+        t_origins = chunk_rays.origins[..., None, :]
+        t_dirs = chunk_rays.viewdirs[..., None, :]
+        positions = t_origins + t_dirs * (t_starts + t_ends) / 2.0
+        return proposal_network(positions)
+
+    def rgb_sigma_fn(t_starts, t_ends):
+        t_origins = chunk_rays.origins[..., None, :]
+        t_dirs = chunk_rays.viewdirs[..., None, :].repeat_interleave(
+            t_starts.shape[-2], dim=-2
+        )
+        positions = t_origins + t_dirs * (t_starts + t_ends) / 2.0
+        return radiance_field(positions, t_dirs)
+
+    results = []
+    chunk = (
+        torch.iinfo(torch.int32).max
+        if radiance_field.training
+        else test_chunk_size
+    )
+    for i in range(0, num_rays, chunk):
+        chunk_rays = namedtuple_map(lambda r: r[i : i + chunk], rays)
+        (
+            rgb,
+            opacity,
+            depth,
+            (weights_per_level, s_vals_per_level),
+        ) = rendering_proposal(
+            rgb_sigma_fn=rgb_sigma_fn,
+            num_samples=num_samples,
+            prop_sigma_fns=[
+                lambda *args: prop_sigma_fn(*args, p) for p in proposal_networks
+            ],
+            num_samples_per_prop=num_samples_per_prop,
+            rays_o=chunk_rays.origins,
+            rays_d=chunk_rays.viewdirs,
+            scene_aabb=scene_aabb,
+            near_plane=near_plane,
+            far_plane=far_plane,
+            stratified=radiance_field.training,
+            sampling_type=sampling_type,
+            opaque_bkgd=opaque_bkgd,
+            render_bkgd=render_bkgd,
+            proposal_requires_grad=proposal_requires_grad,
+            proposal_annealing=proposal_annealing,
+        )
+        chunk_results = [rgb, opacity, depth]
+        results.append(chunk_results)
+
+    colors, opacities, depths = collate(
+        results,
+        collate_fn_map={
+            **default_collate_fn_map,
+            torch.Tensor: lambda x, **_: torch.cat(x, 0),
+        },
+    )
+    return (
+        colors.view((*rays_shape[:-1], -1)),
+        opacities.view((*rays_shape[:-1], -1)),
+        depths.view((*rays_shape[:-1], -1)),
+        weights_per_level,
+        s_vals_per_level,
+    )

nerfacc/__init__.py

 """
 Copyright (c) 2022 Ruilong Li, UC Berkeley.
 """
-import warnings
-
-from .cdf import ray_resampling
 from .contraction import ContractionType, contract, contract_inv
 from .grid import Grid, OccupancyGrid, query_grid
 from .intersection import ray_aabb_intersect
-from .losses import distortion as loss_distortion
 from .pack import pack_data, pack_info, unpack_data, unpack_info
 from .ray_marching import ray_marching
 from .version import __version__
@@ -21,39 +17,30 @@ from .vol_rendering import (
     rendering,
 )
 
-
-# About to be deprecated
-def unpack_to_ray_indices(*args, **kwargs):
-    warnings.warn(
-        "`unpack_to_ray_indices` will be deprecated. Please use `unpack_info` instead.",
-        DeprecationWarning,
-        stacklevel=2,
-    )
-    return unpack_info(*args, **kwargs)
-
-
 __all__ = [
     "__version__",
+    # occ grid
     "Grid",
     "OccupancyGrid",
     "query_grid",
     "ContractionType",
+    # contraction
     "contract",
     "contract_inv",
+    # marching
     "ray_aabb_intersect",
     "ray_marching",
+    # rendering
     "accumulate_along_rays",
     "render_visibility",
     "render_weight_from_alpha",
     "render_weight_from_density",
+    "render_transmittance_from_density",
+    "render_transmittance_from_alpha",
     "rendering",
+    # pack
     "pack_data",
     "unpack_data",
     "unpack_info",
     "pack_info",
-    "ray_resampling",
-    "loss_distortion",
-    "unpack_to_ray_indices",
-    "render_transmittance_from_density",
-    "render_transmittance_from_alpha",
 ]

nerfacc/cdf.py

-"""
-Copyright (c) 2022 Ruilong Li, UC Berkeley.
-"""
-
-from typing import Tuple
-
-from torch import Tensor
-
-import nerfacc.cuda as _C
-
-
-def ray_resampling(
-    packed_info: Tensor,
-    t_starts: Tensor,
-    t_ends: Tensor,
-    weights: Tensor,
-    n_samples: int,
-) -> Tuple[Tensor, Tensor, Tensor]:
-    """Resample a set of rays based on the CDF of the weights.
-
-    Args:
-        packed_info (Tensor): Stores information on which samples belong to the same ray. \
-            See :func:`nerfacc.ray_marching` for details. Tensor with shape (n_rays, 2).
-        t_starts: Where the frustum-shape sample starts along a ray. Tensor with \
-            shape (n_samples, 1).
-        t_ends: Where the frustum-shape sample ends along a ray. Tensor with \
-            shape (n_samples, 1).
-        weights: Volumetric rendering weights for those samples. Tensor with shape \
-            (n_samples,).
-        n_samples (int): Number of samples per ray to resample.
-
-    Returns:
-        Resampled packed info (n_rays, 2), t_starts (n_samples, 1), and t_ends (n_samples, 1).
-    """
-    (
-        resampled_packed_info,
-        resampled_t_starts,
-        resampled_t_ends,
-    ) = _C.ray_resampling(
-        packed_info.contiguous(),
-        t_starts.contiguous(),
-        t_ends.contiguous(),
-        weights.contiguous(),
-        n_samples,
-    )
-    return resampled_packed_info, resampled_t_starts, resampled_t_ends

nerfacc/cuda/__init__.py

@@ -23,7 +23,6 @@ grid_query = _make_lazy_cuda_func("grid_query")
 
 ray_aabb_intersect = _make_lazy_cuda_func("ray_aabb_intersect")
 ray_marching = _make_lazy_cuda_func("ray_marching")
-ray_resampling = _make_lazy_cuda_func("ray_resampling")
 
 is_cub_available = _make_lazy_cuda_func("is_cub_available")
 transmittance_from_sigma_forward_cub = _make_lazy_cuda_func(
@@ -68,3 +67,6 @@ weight_from_alpha_backward_naive = _make_lazy_cuda_func(
 unpack_data = _make_lazy_cuda_func("unpack_data")
 unpack_info = _make_lazy_cuda_func("unpack_info")
 unpack_info_to_mask = _make_lazy_cuda_func("unpack_info_to_mask")
+
+pdf_readout = _make_lazy_cuda_func("pdf_readout")
+pdf_sampling = _make_lazy_cuda_func("pdf_sampling")

nerfacc/cuda/csrc/cdf.cu

-/*
- * Copyright (c) 2022 Ruilong Li, UC Berkeley.
- */
-
-#include "include/helpers_cuda.h"
-
-template <typename scalar_t>
-__global__ void cdf_resampling_kernel(
-    const uint32_t n_rays,
-    const int *packed_info,  // input ray & point indices.
-    const scalar_t *starts,  // input start t
-    const scalar_t *ends,    // input end t
-    const scalar_t *weights, // transmittance weights
-    const int *resample_packed_info,
-    scalar_t *resample_starts,
-    scalar_t *resample_ends)
-{
-    CUDA_GET_THREAD_ID(i, n_rays);
-
-    // locate
-    const int base = packed_info[i * 2 + 0];  // point idx start.
-    const int steps = packed_info[i * 2 + 1]; // point idx shift.
-    const int resample_base = resample_packed_info[i * 2 + 0];
-    const int resample_steps = resample_packed_info[i * 2 + 1];
-    if (steps == 0)
-        return;
-
-    starts += base;
-    ends += base;
-    weights += base;
-    resample_starts += resample_base;
-    resample_ends += resample_base;
-
-    // normalize weights **per ray**
-    scalar_t weights_sum = 0.0f;
-    for (int j = 0; j < steps; j++)
-        weights_sum += weights[j];
-    scalar_t padding = fmaxf(1e-5f - weights_sum, 0.0f);
-    scalar_t padding_step = padding / steps;
-    weights_sum += padding;
-
-    int num_bins = resample_steps + 1;
-    scalar_t cdf_step_size = (1.0f - 1.0 / num_bins) / resample_steps;
-
-    int idx = 0, j = 0;
-    scalar_t cdf_prev = 0.0f, cdf_next = (weights[idx] + padding_step) / weights_sum;
-    scalar_t cdf_u = 1.0 / (2 * num_bins);
-    while (j < num_bins)
-    {
-        if (cdf_u < cdf_next)
-        {
-            // printf("cdf_u: %f, cdf_next: %f\n", cdf_u, cdf_next);
-            // resample in this interval
-            scalar_t scaling = (ends[idx] - starts[idx]) / (cdf_next - cdf_prev);
-            scalar_t t = (cdf_u - cdf_prev) * scaling + starts[idx];
-            if (j < num_bins - 1)
-                resample_starts[j] = t;
-            if (j > 0)
-                resample_ends[j - 1] = t;
-            // going further to next resample
-            cdf_u += cdf_step_size;
-            j += 1;
-        }
-        else
-        {
-            // going to next interval
-            idx += 1;
-            cdf_prev = cdf_next;
-            cdf_next += (weights[idx] + padding_step) / weights_sum;
-        }
-    }
-    if (j != num_bins)
-    {
-        printf("Error: %d %d %f\n", j, num_bins, weights_sum);
-    }
-    return;
-}
-
-// template <typename scalar_t>
-// __global__ void cdf_resampling_kernel(
-//     const uint32_t n_rays,
-//     const int *packed_info,   // input ray & point indices.
-//     const scalar_t *starts,   // input start t
-//     const scalar_t *ends,     // input end t
-//     const scalar_t *weights,  // transmittance weights
-//     const int *resample_packed_info,
-//     scalar_t *resample_starts,
-//     scalar_t *resample_ends)
-// {
-//     CUDA_GET_THREAD_ID(i, n_rays);
-
-//     // locate
-//     const int base = packed_info[i * 2 + 0];  // point idx start.
-//     const int steps = packed_info[i * 2 + 1]; // point idx shift.
-//     const int resample_base = resample_packed_info[i * 2 + 0];
-//     const int resample_steps = resample_packed_info[i * 2 + 1];
-//     if (steps == 0)
-//         return;
-
-//     starts += base;
-//     ends += base;
-//     weights += base;
-//     resample_starts += resample_base;
-//     resample_ends += resample_base;
-
-//     scalar_t cdf_step_size = 1.0f / resample_steps;
-
-//     // normalize weights **per ray**
-//     scalar_t weights_sum = 0.0f;
-//     for (int j = 0; j < steps; j++)
-//         weights_sum += weights[j];
-
-//     scalar_t padding = fmaxf(1e-5f - weights_sum, 0.0f);
-//     scalar_t padding_step = padding / steps;
-//     weights_sum += padding;
-
-//     int idx = 0, j = 0;
-//     scalar_t cdf_prev = 0.0f, cdf_next = (weights[idx] + padding_step) / weights_sum;
-//     scalar_t cdf_u = 0.5f * cdf_step_size;
-//     while (cdf_u < 1.0f)
-//     {
-//         if (cdf_u < cdf_next)
-//         {
-//             // resample in this interval
-//             scalar_t scaling = (ends[idx] - starts[idx]) / (cdf_next - cdf_prev);
-//             scalar_t resample_mid = (cdf_u - cdf_prev) * scaling + starts[idx];
-//             scalar_t resample_half_size = cdf_step_size * scaling * 0.5;
-//             resample_starts[j] = fmaxf(resample_mid - resample_half_size, starts[idx]);
-//             resample_ends[j] = fminf(resample_mid + resample_half_size, ends[idx]);
-//             // going further to next resample
-//             cdf_u += cdf_step_size;
-//             j += 1;
-//         }
-//         else
-//         {
-//             // go to next interval
-//             idx += 1;
-//             if (idx == steps)
-//                 break;
-//             cdf_prev = cdf_next;
-//             cdf_next += (weights[idx] + padding_step) / weights_sum;
-//         }
-//     }
-//     if (j != resample_steps)
-//     {
-//         printf("Error: %d %d %f\n", j, resample_steps, weights_sum);
-//     }
-//     return;
-// }
-
-std::vector<torch::Tensor> ray_resampling(
-    torch::Tensor packed_info,
-    torch::Tensor starts,
-    torch::Tensor ends,
-    torch::Tensor weights,
-    const int steps)
-{
-    DEVICE_GUARD(packed_info);
-
-    CHECK_INPUT(packed_info);
-    CHECK_INPUT(starts);
-    CHECK_INPUT(ends);
-    CHECK_INPUT(weights);
-
-    TORCH_CHECK(packed_info.ndimension() == 2 & packed_info.size(1) == 2);
-    TORCH_CHECK(starts.ndimension() == 2 & starts.size(1) == 1);
-    TORCH_CHECK(ends.ndimension() == 2 & ends.size(1) == 1);
-    TORCH_CHECK(weights.ndimension() == 1);
-
-    const uint32_t n_rays = packed_info.size(0);
-    const uint32_t n_samples = weights.size(0);
-
-    const int threads = 256;
-    const int blocks = CUDA_N_BLOCKS_NEEDED(n_rays, threads);
-
-    torch::Tensor num_steps = torch::split(packed_info, 1, 1)[1];
-    torch::Tensor resample_num_steps = (num_steps > 0).to(num_steps.options()) * steps;
-    torch::Tensor resample_cum_steps = resample_num_steps.cumsum(0, torch::kInt32);
-    torch::Tensor resample_packed_info = torch::cat(
-        {resample_cum_steps - resample_num_steps, resample_num_steps}, 1);
-
-    int total_steps = resample_cum_steps[resample_cum_steps.size(0) - 1].item<int>();
-    torch::Tensor resample_starts = torch::zeros({total_steps, 1}, starts.options());
-    torch::Tensor resample_ends = torch::zeros({total_steps, 1}, ends.options());
-
-    AT_DISPATCH_FLOATING_TYPES_AND_HALF(
-        weights.scalar_type(),
-        "ray_resampling",
-        ([&]
-         { cdf_resampling_kernel<scalar_t><<<blocks, threads, 0, at::cuda::getCurrentCUDAStream()>>>(
-               n_rays,
-               // inputs
-               packed_info.data_ptr<int>(),
-               starts.data_ptr<scalar_t>(),
-               ends.data_ptr<scalar_t>(),
-               weights.data_ptr<scalar_t>(),
-               resample_packed_info.data_ptr<int>(),
-               // outputs
-               resample_starts.data_ptr<scalar_t>(),
-               resample_ends.data_ptr<scalar_t>()); }));
-
-    return {resample_packed_info, resample_starts, resample_ends};
-}