[Enhance] Support nuscenes demo (#353)

* support nuscenes dataset in demo

* add convert_SyncBN in __init__

* fix meshlab visualization bug

* modify meshlab unittest

* add docstring

* add empty line in docstring
Co-authored-by: xiliu8006 <xiliu800@gmail.com>

mmdet3d/apis/__init__.py

-from .inference import inference_detector, init_detector, show_result_meshlab
+from .inference import (convert_SyncBN, inference_detector, init_detector,
+                        show_result_meshlab)
 from .test import single_gpu_test
 
 __all__ = [
     'inference_detector', 'init_detector', 'single_gpu_test',
-    'show_result_meshlab'
+    'show_result_meshlab', 'convert_SyncBN'
 ]

mmdet3d/apis/inference.py

@@ -11,6 +11,22 @@ from mmdet3d.datasets.pipelines import Compose
 from mmdet3d.models import build_detector
 
 
+def convert_SyncBN(config):
+    """Convert config's naiveSyncBN to BN.
+
+    Args:
+         config (str or :obj:`mmcv.Config`): Config file path or the config
+            object.
+    """
+    if isinstance(config, dict):
+        for item in config:
+            if item == 'norm_cfg':
+                config[item]['type'] = config[item]['type']. \
+                                    replace('naiveSyncBN', 'BN')
+            else:
+                convert_SyncBN(config[item])
+
+
 def init_detector(config, checkpoint=None, device='cuda:0'):
     """Initialize a detector from config file.
 
@@ -30,6 +46,7 @@ def init_detector(config, checkpoint=None, device='cuda:0'):
         raise TypeError('config must be a filename or Config object, '
                         f'but got {type(config)}')
     config.model.pretrained = None
+    convert_SyncBN(config.model)
     config.model.train_cfg = None
     model = build_detector(config.model, test_cfg=config.get('test_cfg'))
     if checkpoint is not None:
@@ -64,6 +81,9 @@ def inference_detector(model, pcd):
         pts_filename=pcd,
         box_type_3d=box_type_3d,
         box_mode_3d=box_mode_3d,
+        sweeps=[],
+        # set timestamp = 0
+        timestamp=[0],
         img_fields=[],
         bbox3d_fields=[],
         pts_mask_fields=[],
@@ -100,6 +120,9 @@ def show_result_meshlab(data, result, out_dir):
 
     assert out_dir is not None, 'Expect out_dir, got none.'
 
+    if 'pts_bbox' in result[0].keys():
+        pred_bboxes = result[0]['pts_bbox']['boxes_3d'].tensor.numpy()
+    else:
         pred_bboxes = result[0]['boxes_3d'].tensor.numpy()
     # for now we convert points into depth mode
     if data['img_metas'][0][0]['box_mode_3d'] != Box3DMode.DEPTH:
@@ -108,7 +131,5 @@ def show_result_meshlab(data, result, out_dir):
         pred_bboxes = Box3DMode.convert(pred_bboxes,
                                         data['img_metas'][0][0]['box_mode_3d'],
                                         Box3DMode.DEPTH)
-        pred_bboxes[..., 2] += pred_bboxes[..., 5] / 2
-    else:
-        pred_bboxes[..., 2] += pred_bboxes[..., 5] / 2
-    show_result(points, None, pred_bboxes, out_dir, file_name)
+    show_result(points, None, pred_bboxes, out_dir, file_name, show=False)
+    return out_dir, file_name

mmdet3d/core/visualizer/show_result.py

@@ -79,9 +79,9 @@ def show_result(points, gt_bboxes, pred_bboxes, out_dir, filename, show=True):
         filename (str): Filename of the current frame.
         show (bool): Visualize the results online.
     """
+    if show:
         from .open3d_vis import Visualizer
 
-    if show:
         vis = Visualizer(points)
         if pred_bboxes is not None:
             vis.add_bboxes(bbox3d=pred_bboxes)

tests/test_runtime/test_apis.py

+import numpy as np
+import os
 import pytest
+import tempfile
 import torch
 from mmcv.parallel import MMDataParallel
 from os.path import dirname, exists, join
 
-from mmdet3d.apis import inference_detector, init_detector, single_gpu_test
+from mmdet3d.apis import (convert_SyncBN, inference_detector, init_detector,
+                          show_result_meshlab, single_gpu_test)
+from mmdet3d.core import Box3DMode
+from mmdet3d.core.bbox import LiDARInstance3DBoxes
 from mmdet3d.datasets import build_dataloader, build_dataset
 from mmdet3d.models import build_detector
 
@@ -32,6 +38,48 @@ def _get_config_module(fname):
     return config_mod
 
 
+def test_convert_SyncBN():
+    cfg = _get_config_module(
+        'pointpillars/hv_pointpillars_fpn_sbn-all_4x8_2x_nus-3d.py')
+    model_cfg = cfg.model
+    convert_SyncBN(model_cfg)
+    assert model_cfg['pts_voxel_encoder']['norm_cfg']['type'] == 'BN1d'
+    assert model_cfg['pts_backbone']['norm_cfg']['type'] == 'BN2d'
+    assert model_cfg['pts_neck']['norm_cfg']['type'] == 'BN2d'
+
+
+def test_show_result_meshlab():
+    pcd = 'tests/data/nuscenes/samples/LIDAR_TOP/n015-2018-08-02-17-16-37+' \
+              '0800__LIDAR_TOP__1533201470948018.pcd.bin'
+    box_3d = LiDARInstance3DBoxes(
+        torch.tensor(
+            [[8.7314, -1.8559, -1.5997, 0.4800, 1.2000, 1.8900, 0.0100]]))
+    labels_3d = torch.tensor([0])
+    scores_3d = torch.tensor([0.5])
+    points = np.random.rand(100, 4)
+    img_meta = dict(
+        pts_filename=pcd, boxes_3d=box_3d, box_mode_3d=Box3DMode.LIDAR)
+    data = dict(points=[[torch.tensor(points)]], img_metas=[[img_meta]])
+    result = [
+        dict(
+            pts_bbox=dict(
+                boxes_3d=box_3d, labels_3d=labels_3d, scores_3d=scores_3d))
+    ]
+    temp_out_dir = tempfile.mkdtemp()
+    out_dir, file_name = show_result_meshlab(data, result, temp_out_dir)
+    expected_outfile_ply = file_name + '_pred.ply'
+    expected_outfile_obj = file_name + '_points.obj'
+    expected_outfile_ply_path = os.path.join(out_dir, file_name,
+                                             expected_outfile_ply)
+    expected_outfile_obj_path = os.path.join(out_dir, file_name,
+                                             expected_outfile_obj)
+    assert os.path.exists(expected_outfile_ply_path)
+    assert os.path.exists(expected_outfile_obj_path)
+    os.remove(expected_outfile_obj_path)
+    os.remove(expected_outfile_ply_path)
+    os.removedirs(os.path.join(temp_out_dir, file_name))
+
+
 def test_inference_detector():
     pcd = 'tests/data/kitti/training/velodyne_reduced/000000.bin'
     detector_cfg = 'configs/pointpillars/hv_pointpillars_secfpn_' \