[Feature] Add ImVoteNet model zoo (#384)

* add model zoo

* change bibtex and more doc

README.md

@@ -86,6 +86,7 @@ Support methods
 - [x] [MVXNet (ICRA'2019)](configs/mvxnet/README.md)
 - [x] [CenterPoint (CVPR'2021)](configs/centerpoint/README.md)
 - [x] [SSN (ECCV'2020)](configs/ssn/README.md)
+- [x] [ImVoteNet (CVPR'2020)](configs/imvotenet/README.md)
 
 |                    | ResNet   | ResNeXt  | SENet    |PointNet++ | HRNet | RegNetX | Res2Net |
 |--------------------|:--------:|:--------:|:--------:|:---------:|:-----:|:--------:|:-----:|
@@ -99,6 +100,7 @@ Support methods
 | MVXNet             | ☐        | ☐        | ☐        | ✗         | ☐     | ✓        | ☐     |
 | CenterPoint        | ☐        | ☐        | ☐        | ✗         | ☐     | ✓        | ☐     |
 | SSN                | ☐        | ☐        | ☐        | ✗         | ☐     | ✓        | ☐     |
+| ImVoteNet            | ✗        | ✗        | ✗        | ✓         | ✗     | ✗        | ✗     |
 
 Other features
 - [x] [Dynamic Voxelization](configs/dynamic_voxelization/README.md)

README_zh-CN.md

@@ -85,6 +85,7 @@ MMDetection3D 是一个基于 PyTorch 的目标检测开源工具箱, 下一代
 - [x] [MVXNet (ICRA'2019)](configs/mvxnet/README.md)
 - [x] [CenterPoint (CVPR'2021)](configs/centerpoint/README.md)
 - [x] [SSN (ECCV'2020)](configs/ssn/README.md)
+- [x] [ImVoteNet (CVPR'2020)](configs/imvotenet/README.md)
 
 |                    | ResNet   | ResNeXt  | SENet    |PointNet++ | HRNet | RegNetX | Res2Net |
 |--------------------|:--------:|:--------:|:--------:|:---------:|:-----:|:--------:|:-----:|
@@ -98,6 +99,7 @@ MMDetection3D 是一个基于 PyTorch 的目标检测开源工具箱, 下一代
 | MVXNet             | ☐        | ☐        | ☐        | ✗         | ☐     | ✓        | ☐     |
 | CenterPoint        | ☐        | ☐        | ☐        | ✗         | ☐     | ✓        | ☐     |
 | SSN                | ☐        | ☐        | ☐        | ✗         | ☐     | ✓        | ☐     |
+| ImVoteNet            | ✗        | ✗        | ✗        | ✓         | ✗     | ✗        | ✗     |
 
 其他特性
 - [x] [Dynamic Voxelization](configs/dynamic_voxelization/README.md)

configs/imvotenet/README.md

+# ImVoteNet: Boosting 3D Object Detection in Point Clouds with Image Votes
+
+## Introduction
+
+[ALGORITHM]
+
+We implement ImVoteNet and provide the result and checkpoints on SUNRGBD.
+
+```
+@inproceedings{qi2020imvotenet,
+  title={Imvotenet: Boosting 3D object detection in point clouds with image votes},
+  author={Qi, Charles R and Chen, Xinlei and Litany, Or and Guibas, Leonidas J},
+  booktitle={Proceedings of the IEEE/CVF conference on computer vision and pattern recognition},
+  pages={4404--4413},
+  year={2020}
+}
+```
+
+## Results
+
+### SUNRGBD-2D (Stage 1, image branch pre-train)
+
+|  Backbone   | Lr schd | Mem (GB) | Inf time (fps) | AP@0.25 |AP@0.5| Download |
+| :---------: | :-----: | :------: | :------------: | :----: |:----: | :------: |
+|    [PointNet++](./imvotenet_faster_rcnn_r50_fpn_2x4_sunrgbd-3d-10class.py)     |   |2.1| ||62.70|[model](https://download.openmmlab.com/mmdetection3d/v0.1.0_models/imvotenet/imvotenet_faster_rcnn_r50_fpn_2x4_sunrgbd-3d-10class/imvotenet_faster_rcnn_r50_fpn_2x4_sunrgbd-3d-10class_20210323_173222-cad62aeb.pth) | [log](https://download.openmmlab.com/mmdetection3d/v0.1.0_models/imvotenet/imvotenet_faster_rcnn_r50_fpn_2x4_sunrgbd-3d-10class/imvotenet_faster_rcnn_r50_fpn_2x4_sunrgbd-3d-10class_20210323_173222.log.json)|
+
+### SUNRGBD-3D (Stage 2)
+
+|  Backbone   | Lr schd | Mem (GB) | Inf time (fps) | AP@0.25 |AP@0.5| Download |
+| :---------: | :-----: | :------: | :------------: | :----: |:----: | :------: |
+|    [PointNet++](./imvotenet_stage2_16x8_sunrgbd-3d-10class.py)     |  3x    |9.4| |64.04||[model](https://download.openmmlab.com/mmdetection3d/v0.1.0_models/imvotenet/imvotenet_stage2_16x8_sunrgbd-3d-10class/imvotenet_stage2_16x8_sunrgbd-3d-10class_20210323_184021-d44dcb66.pth) | [log](https://download.openmmlab.com/mmdetection3d/v0.1.0_models/imvotenet/imvotenet_stage2_16x8_sunrgbd-3d-10class/imvotenet_stage2_16x8_sunrgbd-3d-10class_20210323_184021.log.json)|

configs/imvotenet/imvotenet_stage2_16x8_sunrgbd-3d-10class.py

@@ -239,4 +239,5 @@ data = dict(
     val=dict(pipeline=test_pipeline),
     test=dict(pipeline=test_pipeline))
 
-load_from = None  # TODO after we update model zoo
+# may also use your own pre-trained image branch
+load_from = 'https://download.openmmlab.com/mmdetection3d/v0.1.0_models/imvotenet/imvotenet_faster_rcnn_r50_fpn_2x4_sunrgbd-3d-10class/imvotenet_faster_rcnn_r50_fpn_2x4_sunrgbd-3d-10class_20210323_173222-cad62aeb.pth'  # noqa

docs/model_zoo.md

@@ -53,3 +53,7 @@ Please refer to [CenterPoint](https://github.com/open-mmlab/mmdetection3d/blob/m
 ### SSN
 
 Please refer to [SSN](https://github.com/open-mmlab/mmdetection3d/blob/master/configs/ssn) for details. We provide pointpillars with shape-aware grouping heads used in SSN on the nuScenes and Lyft dataset currently.
+
+### ImVoteNet
+
+Please refer to [ImVoteNet](https://github.com/open-mmlab/mmdetection3d/blob/master/configs/imvotenet) for details. We provide VoteNet baselines on SUNRGBD dataset.