Skip to content

GitLab

Commit d2b71343 authored by 雍大凯's avatar 雍大凯
Browse files

add code

parent 69e57885
Hide whitespace changes
Inline Side-by-side
Showing with 1111 additions and 6 deletions
BEVFormer @ 66b65f3a
Subproject commit 66b65f3a1f58caf0507cb2a971b9c0e7f842376c
docker-hub/BEVFormer/readme.md
View file @ d2b71343
# BEVformer
## 环境构建
## 快速执行
### 环境构建
```
docker run -dit --network=host --name=bevformer --privileged --device=/dev/kfd --device=/dev/dri --ipc=host --shm-size=128G --group-add video --cap-add=SYS_PTRACE --security-opt seccomp=unconfined -u root --ulimit stack=-1:-1 --ulimit memlock=-1:-1 -v /opt/hyhal:/opt/hyhal:ro -v /public/opendas/DL_DATA/zijia:/data:ro harbor.sourcefind.cn:5443/dcu/admin/base/custom:pytorch2.5.1-driver-bevformer
docker run -dit --network=host --name=bevformer --privileged --device=/dev/kfd --device=/dev/dri --ipc=host --shm-size=128G --group-add video --cap-add=SYS_PTRACE --security-opt seccomp=unconfined -u root --ulimit stack=-1:-1 --ulimit memlock=-1:-1 -v /opt/hyhal:/opt/hyhal:ro -v /public/opendas/DL_DATA/zijia:/data:ro harbor.sourcefind.cn:5443/dcu/admin/base/custom:pytorch2.5.1-driver-bevformer
```
数据集已完成处理在乌镇集群/public/opendas/DL_DATA/zijia路径内
## 代码执行
### 代码执行
```
## 代码已经在容器/workspace下了
cd /workspace/BEVFormer
bash start_bevformer.sh
```
docker-hub/FlashOCC/Flashocc/LICENSE 0 → 100644
View file @ d2b71343
Apache License
Version 2.0, January 2004
http://www.apache.org/licenses/
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
1. Definitions.
"License" shall mean the terms and conditions for use, reproduction,
and distribution as defined by Sections 1 through 9 of this document.
"Licensor" shall mean the copyright owner or entity authorized by
the copyright owner that is granting the License.
"Legal Entity" shall mean the union of the acting entity and all
other entities that control, are controlled by, or are under common
control with that entity. For the purposes of this definition,
"control" means (i) the power, direct or indirect, to cause the
direction or management of such entity, whether by contract or
otherwise, or (ii) ownership of fifty percent (50%) or more of the
outstanding shares, or (iii) beneficial ownership of such entity.
"You" (or "Your") shall mean an individual or Legal Entity
exercising permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation
source, and configuration files.
"Object" form shall mean any form resulting from mechanical
transformation or translation of a Source form, including but
not limited to compiled object code, generated documentation,
and conversions to other media types.
"Work" shall mean the work of authorship, whether in Source or
Object form, made available under the License, as indicated by a
copyright notice that is included in or attached to the work
(an example is provided in the Appendix below).
"Derivative Works" shall mean any work, whether in Source or Object
form, that is based on (or derived from) the Work and for which the
editorial revisions, annotations, elaborations, or other modifications
represent, as a whole, an original work of authorship. For the purposes
of this License, Derivative Works shall not include works that remain
separable from, or merely link (or bind by name) to the interfaces of,
the Work and Derivative Works thereof.
"Contribution" shall mean any work of authorship, including
the original version of the Work and any modifications or additions
to that Work or Derivative Works thereof, that is intentionally
submitted to Licensor for inclusion in the Work by the copyright owner
or by an individual or Legal Entity authorized to submit on behalf of
the copyright owner. For the purposes of this definition, "submitted"
means any form of electronic, verbal, or written communication sent
to the Licensor or its representatives, including but not limited to
communication on electronic mailing lists, source code control systems,
and issue tracking systems that are managed by, or on behalf of, the
Licensor for the purpose of discussing and improving the Work, but
excluding communication that is conspicuously marked or otherwise
designated in writing by the copyright owner as "Not a Contribution."
"Contributor" shall mean Licensor and any individual or Legal Entity
on behalf of whom a Contribution has been received by Licensor and
subsequently incorporated within the Work.
2. Grant of Copyright License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
copyright license to reproduce, prepare Derivative Works of,
publicly display, publicly perform, sublicense, and distribute the
Work and such Derivative Works in Source or Object form.
3. Grant of Patent License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
(except as stated in this section) patent license to make, have made,
use, offer to sell, sell, import, and otherwise transfer the Work,
where such license applies only to those patent claims licensable
by such Contributor that are necessarily infringed by their
Contribution(s) alone or by combination of their Contribution(s)
with the Work to which such Contribution(s) was submitted. If You
institute patent litigation against any entity (including a
cross-claim or counterclaim in a lawsuit) alleging that the Work
or a Contribution incorporated within the Work constitutes direct
or contributory patent infringement, then any patent licenses
granted to You under this License for that Work shall terminate
as of the date such litigation is filed.
4. Redistribution. You may reproduce and distribute copies of the
Work or Derivative Works thereof in any medium, with or without
modifications, and in Source or Object form, provided that You
meet the following conditions:
(a) You must give any other recipients of the Work or
Derivative Works a copy of this License; and
(b) You must cause any modified files to carry prominent notices
stating that You changed the files; and
(c) You must retain, in the Source form of any Derivative Works
that You distribute, all copyright, patent, trademark, and
attribution notices from the Source form of the Work,
excluding those notices that do not pertain to any part of
the Derivative Works; and
(d) If the Work includes a "NOTICE" text file as part of its
distribution, then any Derivative Works that You distribute must
include a readable copy of the attribution notices contained
within such NOTICE file, excluding those notices that do not
pertain to any part of the Derivative Works, in at least one
of the following places: within a NOTICE text file distributed
as part of the Derivative Works; within the Source form or
documentation, if provided along with the Derivative Works; or,
within a display generated by the Derivative Works, if and
wherever such third-party notices normally appear. The contents
of the NOTICE file are for informational purposes only and
do not modify the License. You may add Your own attribution
notices within Derivative Works that You distribute, alongside
or as an addendum to the NOTICE text from the Work, provided
that such additional attribution notices cannot be construed
as modifying the License.
You may add Your own copyright statement to Your modifications and
may provide additional or different license terms and conditions
for use, reproduction, or distribution of Your modifications, or
for any such Derivative Works as a whole, provided Your use,
reproduction, and distribution of the Work otherwise complies with
the conditions stated in this License.
5. Submission of Contributions. Unless You explicitly state otherwise,
any Contribution intentionally submitted for inclusion in the Work
by You to the Licensor shall be under the terms and conditions of
this License, without any additional terms or conditions.
Notwithstanding the above, nothing herein shall supersede or modify
the terms of any separate license agreement you may have executed
with Licensor regarding such Contributions.
6. Trademarks. This License does not grant permission to use the trade
names, trademarks, service marks, or product names of the Licensor,
except as required for reasonable and customary use in describing the
origin of the Work and reproducing the content of the NOTICE file.
7. Disclaimer of Warranty. Unless required by applicable law or
agreed to in writing, Licensor provides the Work (and each
Contributor provides its Contributions) on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied, including, without limitation, any warranties or conditions
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
PARTICULAR PURPOSE. You are solely responsible for determining the
appropriateness of using or redistributing the Work and assume any
risks associated with Your exercise of permissions under this License.
8. Limitation of Liability. In no event and under no legal theory,
whether in tort (including negligence), contract, or otherwise,
unless required by applicable law (such as deliberate and grossly
negligent acts) or agreed to in writing, shall any Contributor be
liable to You for damages, including any direct, indirect, special,
incidental, or consequential damages of any character arising as a
result of this License or out of the use or inability to use the
Work (including but not limited to damages for loss of goodwill,
work stoppage, computer failure or malfunction, or any and all
other commercial damages or losses), even if such Contributor
has been advised of the possibility of such damages.
9. Accepting Warranty or Additional Liability. While redistributing
the Work or Derivative Works thereof, You may choose to offer,
and charge a fee for, acceptance of support, warranty, indemnity,
or other liability obligations and/or rights consistent with this
License. However, in accepting such obligations, You may act only
on Your own behalf and on Your sole responsibility, not on behalf
of any other Contributor, and only if You agree to indemnify,
defend, and hold each Contributor harmless for any liability
incurred by, or claims asserted against, such Contributor by reason
of your accepting any such warranty or additional liability.
END OF TERMS AND CONDITIONS
APPENDIX: How to apply the Apache License to your work.
To apply the Apache License to your work, attach the following
boilerplate notice, with the fields enclosed by brackets "[]"
replaced with your own identifying information. (Don't include
the brackets!) The text should be enclosed in the appropriate
comment syntax for the file format. We also recommend that a
file or class name and description of purpose be included on the
same "printed page" as the copyright notice for easier
identification within third-party archives.
Copyright (c) 2024 Institute of Intelligent Control, Dalian University of Technology. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
docker-hub/FlashOCC/Flashocc/README.md 0 → 100644
View file @ d2b71343
# FlashOcc: Fast and Memory-Efficient Occupancy Prediction via Channel-to-Height Plugin
<div align="left">
<img src="figs/performance_flashocc.jpg" width="1500px" />
</div><br/>
* Please note that the FPS here is measured with RTX3090 TensorRT FP16.
# Panoptic-FlashOcc: An Efficient Baseline to Marry Semantic Occupancy with Panoptic via Instance Center
<div align="left">
<img src="figs/performance.png" width="1500px"/>
</div><br/>
* Please note that the FPS here is measured with A100 GPU (PyTorch fp32 backend).
## News
- **2024.09.16** Technical Report: FlashOcc can be insert to Bevdet with 1.1ms consumption while facilitating each other.[![arXiv](https://img.shields.io/badge/arXiv-Paper-<COLOR>.svg)](https://arxiv.org/abs/2409.11160)
- **2024.09.16** [Selected as reference algorithm for occupancy on horizon J6E/M](https://zhuanlan.zhihu.com/p/720461546)
- **2024.06.10** Release the code for Panoptic-FlashOCC
- **2024.04.17** Support for ray-iou metric
- **2024.03.22** Release the code for FlashOCCV2
- **2024.02.03** [Release the training code for FlashOcc on UniOcc](https://github.com/drilistbox/FlashOCC_on_UniOcc_and_RenderOCC)
- **2024.01.20** [TensorRT Implement Writen In C++ With Cuda Acceleration](https://github.com/drilistbox/TRT_FlashOcc)
- **2023.12.23** Release the quick testing code via TensorRT in MMDeploy.
- **2023.11.28** Release the training code for FlashOCC.
<!-- - [History](./docs/en/news.md) -->
<!-- ## Introduction -->
[![arXiv](https://img.shields.io/badge/arXiv-Paper-<COLOR>.svg)](https://arxiv.org/abs/2311.12058)
[![arXiv](https://img.shields.io/badge/arXiv-Paper-<COLOR>.svg)](https://arxiv.org/pdf/2406.10527)
[![arXiv](https://img.shields.io/badge/arXiv-Paper-<COLOR>.svg)](https://arxiv.org/abs/2409.11160)
This repository is an official implementation of [FlashOCC](https://arxiv.org/abs/2311.12058)
<div align="center">
<img src="figs/overview.png"/>
</div><br/>
and [Panoptic-FlashOCC](https://arxiv.org/pdf/2406.10527)
<div align="center">
<img src="figs/panoptic_flashOcc.jpg"/>
</div><br/>
## Main Results
### 1. FlashOCC
| Config | Backbone | Input <br/>Size | mIoU | FPS<br/>(Hz) | Flops<br/>(G) | Params<br/>(M) | Model | Log |
|:---------------------------------------------------------------------------------------------------|:--------:|:----------:|:-------:|:-----------------------------:|:------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------:|:-------------------------------------------------------------------:|:----------------------------------------------------------------------------------------------:|
| [**BEVDetOCC (1f)**](projects/configs/bevdet_occ/bevdet-occ-r50.py) | R50 | 256x704 | 31.60 | 92.1 | [241.76](doc/mmdeploy_test.md) | [29.02](doc/mmdeploy_test.md) | [gdrive]() | [log]() |
| [**M0: FlashOCC (1f)**](projects/configs/flashocc/flashocc-r50.py) | R50 | 256x704 | 31.95 | [197.6](doc/mmdeploy_test.md) | [154.1](doc/mmdeploy_test.md) | [39.94](doc/mmdeploy_test.md) | [gdrive](https://drive.google.com/file/d/14my3jdqiIv6VIrkozQ6-ruEcBOPVlWGJ/view?usp=sharing) | [log](https://drive.google.com/file/d/1E-kaHxbTr6s3Qn70vhKpwJM8kejoNFxQ/view?usp=sharing) |
| [**M1: FlashOCC (1f)**](projects/configs/flashocc/flashocc-r50.py) | R50 | 256x704 | 32.08 | [152.7](doc/mmdeploy_test.md) | [248.57](doc/mmdeploy_test.md) | [44.74](doc/mmdeploy_test.md) | [gdrive](https://drive.google.com/file/d/1k9BzXB2nRyvXhqf7GQx3XNSej6Oq6I-B/view?usp=drive_link) | [log](https://drive.google.com/file/d/1NRm27wVZMSUylmZxsMedFSLr7729YEAV/view?usp=drive_link) |
| [**BEVDetOCC-4D-Stereo (2f)**](projects/configs/bevdet_occ/bevdet-occ-r50-4d-stereo.py) | R50 | 256x704 | 36.1 | - | - | - | [baidu](https://pan.baidu.com/s/1237QyV18zvRJ1pU3YzRItw?pwd=npe1) | [log](https://pan.baidu.com/s/1237QyV18zvRJ1pU3YzRItw?pwd=npe1) |
| [**M2:FlashOCC-4D-Stereo (2f)**](projects/configs/flashocc/flashocc-r50-4d-stereo.py) | R50 | 256x704 | 37.84 | - | - | - | [gdrive](https://drive.google.com/file/d/12WYaCdoZA8-A6_oh6vdLgOmqyEc3PNCe/view?usp=drive_link) | [log](https://drive.google.com/file/d/1eYvu9gUSQ7qk7w7lWPLrZMB0O2uKQUk3/view?usp=drive_link) |
| [**BEVDetOCC-4D-Stereo (2f)**](projects/configs/bevdet_occ/bevdet-occ-stbase-4d-stereo-512x1408.py) | Swin-B | 512x1408 | 42.0 | - | - | - | [baidu](https://pan.baidu.com/s/1237QyV18zvRJ1pU3YzRItw?pwd=npe1) | [log](https://pan.baidu.com/s/1237QyV18zvRJ1pU3YzRItw?pwd=npe1) |
| [**M3:FlashOCC-4D-Stereo (2f)**](projects/configs/flashocc/flashocc-stbase-4d-stereo-512x1408_4x4_2e-4.py) | Swin-B | 512x1408 | 43.52 | - | [1490.77](doc/mmdeploy_test.md) | [144.99](doc/mmdeploy_test.md) | [gdrive](https://drive.google.com/file/d/1f6E6Bm6enIJETSEbfXs57M0iOUU997kU/view?usp=drive_link) | [log](https://drive.google.com/file/d/1tch-YK4ROGDGNmDcN5FZnOAvsbHe-iSU/view?usp=drive_link) |
FPS are tested via TensorRT on 3090 with FP16 precision. Please refer to Tab.2 in paper for the detail model settings for M-number.
### 2. Panoptic-FlashOCC
**In Panoptic-FlashOCC, we have made the following 3 adjustments to FlashOCC**:
- Without using camera mask for training. This is because its use significantly improves the prediction performance in the visible region, but at the expense of prediction in the invisible region.
- Using category balancing.
- Using stronger loss settings.
- Introducing instance center for panoptic occupancy
**More results for different configurations will be released soon.**
| Config | Backbone | Input <br/>Size | RayIou | RayPQ | mIoU | FPS<br/>(Hz) | Flops<br/>(G) | Params<br/>(M) | Model | Log |
|:--------------------------------------------------------------------------------|:--------:|:-----------:|:-------:|:------:|:------:|:------------:|:------------------------------:|:-----------------------------:|:-----------------------------------------------------------------------------------------------:|:--------------------------------------------------------------------------------------------:|
| [**M1: FlashOCC (1f)**](projects/configs/flashocc/flashocc-r50.py) | R50 | 256x704 | - | - | 15.41 | - | [248.57](doc/mmdeploy_test.md) | [44.74](doc/mmdeploy_test.md) | [gdrive](https://drive.google.com/file/d/14XsvjSwp-vLpy_eBZzvKo3MAh-YWRHcu/view?usp=drive_link) | [log](https://drive.google.com/file/d/1cTDoauEmjhK9fReLDcA2zPZx4a6X3U1-/view?usp=drive_link) |
| [**Panoptic-FlashOCC-Depth-tiny (1f)**](projects/configs/panoptic-flashocc/panoptic-flashocc-r50-depth-tiny.py) | R50 | 256x704 | 34.57 | - | 28.83 | 43.9 | [175.00](doc/mmdeploy_test.md) | [45.32](doc/mmdeploy_test.md) | [gdrive](https://drive.google.com/drive/folders/1cgCsbXgikoP10lj6DBC7Le9C-UOCIxlN?usp=sharing) | [log](https://drive.google.com/drive/folders/1cgCsbXgikoP10lj6DBC7Le9C-UOCIxlN?usp=sharing) |
| [**Panoptic-FlashOCC-Depth-tiny-Pano (1f)**](projects/configs/panoptic-flashocc/panoptic-flashocc-r50-depth-tiny-pano.py) | R50 | 256x704 | 34.81 | 12.9 | 29.14 | 39.8 | [175.00](doc/mmdeploy_test.md) | [45.32](doc/mmdeploy_test.md) | [gdrive](https://drive.google.com/drive/folders/1cgCsbXgikoP10lj6DBC7Le9C-UOCIxlN?usp=sharing) | [log](https://drive.google.com/drive/folders/1cgCsbXgikoP10lj6DBC7Le9C-UOCIxlN?usp=sharing) |
| [**Panoptic-FlashOCC-Depth (1f)**](projects/configs/panoptic-flashocc/panoptic-flashocc-r50-depth.py) | R50 | 256x704 | 34.93 | - | 28.91 | 38.7 | [269.47](doc/mmdeploy_test.md) | [50.12](doc/mmdeploy_test.md) | [gdrive](https://drive.google.com/drive/folders/1cgCsbXgikoP10lj6DBC7Le9C-UOCIxlN?usp=sharing) | [log](https://drive.google.com/drive/folders/1cgCsbXgikoP10lj6DBC7Le9C-UOCIxlN?usp=sharing) |
| [**Panoptic-FlashOCC-Depth-Pano (1f)**](projects/configs/panoptic-flashocc/panoptic-flashocc-r50-depth-pano.py) | R50 | 256x704 | 35.22 | 13.2 | 29.39 | 35.2 | [269.47](doc/mmdeploy_test.md) | [50.12](doc/mmdeploy_test.md) | [gdrive](https://drive.google.com/drive/folders/1cgCsbXgikoP10lj6DBC7Le9C-UOCIxlN?usp=sharing) | [log](https://drive.google.com/drive/folders/1cgCsbXgikoP10lj6DBC7Le9C-UOCIxlN?usp=sharing) |
| [**Panoptic-FlashOCC-4D-Depth (2f)**](projects/configs/panoptic-flashocc/panoptic-flashocc-r50-depth4d.py) | R50 | 256x704 | 35.99 | - | 29.57 | 35.9 | - | - | [gdrive](https://drive.google.com/drive/folders/1cgCsbXgikoP10lj6DBC7Le9C-UOCIxlN?usp=sharing) | [log](https://drive.google.com/drive/folders/1cgCsbXgikoP10lj6DBC7Le9C-UOCIxlN?usp=sharing) |
| [**Panoptic-FlashOCC-4D-Depth-Pano (2f)**](projects/configs/panoptic-flashocc/panoptic-flashocc-r50-depth4d-pano.py) | R50 | 256x704 | 36.76 | 14.5 | 30.31 | 30.4 | - | - | [gdrive](https://drive.google.com/drive/folders/1cgCsbXgikoP10lj6DBC7Le9C-UOCIxlN?usp=sharing) | [log](https://drive.google.com/drive/folders/1cgCsbXgikoP10lj6DBC7Le9C-UOCIxlN?usp=sharing) |
| [**Panoptic-FlashOCC-4DLongterm-Depth (8f)**](projects/configs/panoptic-flashocc/panoptic-flashocc-r50-depth4d-longterm8f.py) | R50 | 256x704 | 38.51 | - | 31.49 | 35.6 | - | - | [gdrive](https://drive.google.com/drive/folders/1cgCsbXgikoP10lj6DBC7Le9C-UOCIxlN?usp=sharing) | [log](https://drive.google.com/drive/folders/1cgCsbXgikoP10lj6DBC7Le9C-UOCIxlN?usp=sharing) |
| [**Panoptic-FlashOCC-4DLongterm-Depth-Pano (8f)**](projects/configs/panoptic-flashocc/panoptic-flashocc-r50-depth4d-longterm8f.py) | R50 | 256x704 | 38.50 | 16.0 | 31.57 | 30.2 | - | - | [gdrive](https://drive.google.com/drive/folders/1cgCsbXgikoP10lj6DBC7Le9C-UOCIxlN?usp=sharing) | [log](https://drive.google.com/drive/folders/1cgCsbXgikoP10lj6DBC7Le9C-UOCIxlN?usp=sharing) |
* Please note that the FPS here is measured with A100 GPU (PyTorch fp32 backend).
## Get Started
1. [Environment Setup](doc/install.md)
2. [Model Training](doc/model_training.md)
3. [Quick Test Via TensorRT In MMDeploy](doc/mmdeploy_test.md)
| Backend | mIOU | FPS(Hz) |
|----------|-------|---------|
| PyTorch-FP32 | 31.95 | - |
| TRT-FP32 | 30.78 | 96.2 |
| TRT-FP16 | 30.78 | 197.6 |
| TRT-FP16+INT8(PTQ) | 29.60 | 383.7 |
| TRT-INT8(PTQ) | 29.59 | 397.0 |
4. [Visualization](doc/visualization.md)
* [flashocc] : A detail video can be found at [baidu](https://pan.baidu.com/s/1xfnFsj5IclpjJxIaOlI6dA?pwd=gype)
<div align="center">
<img src="figs/visualization.png"/>
</div><br/>
* [panoptic-flashocc] : first row is our prediction and second row is gt.
<div align="center">
<img src="figs/sem.png"/>
</div><br/>
<div align="center">
<img src="figs/pano.png"/>
</div><br/>
5. [TensorRT Implement Writen In C++ With Cuda Acceleration](https://github.com/drilistbox/TRT_FlashOcc)
## Acknowledgement
Many thanks to the authors of [BEVDet](https://github.com/HuangJunJie2017/BEVDet), [FB-BEV](https://github.com/NVlabs/FB-BEV.git),
[RenderOcc](https://github.com/pmj110119/RenderOcc.git) and [SparseBEV](https://github.com/MCG-NJU/SparseBEV.git)
## Bibtex
If this work is helpful for your research, please consider citing the following BibTeX entry.
```
@article{yu2024ultimatedo,
title={UltimateDO: An Efficient Framework to Marry Occupancy Prediction with 3D Object Detection via Channel2height},
author={Yu, Zichen and Shu, Changyong},
journal={arXiv preprint arXiv:2409.11160},
year={2024}
}
@article{yu2024panoptic,
title={Panoptic-FlashOcc: An Efficient Baseline to Marry Semantic Occupancy with Panoptic via Instance Center},
author={Yu, Zichen and Shu, Changyong and Sun, Qianpu and Linghu, Junjie and Wei, Xiaobao and Yu, Jiangyong and Liu, Zongdai and Yang, Dawei and Li, Hui and Chen, Yan},
journal={arXiv preprint arXiv:2406.10527},
year={2024}
}
@article{yu2023flashocc,
title={FlashOcc: Fast and Memory-Efficient Occupancy Prediction via Channel-to-Height Plugin},
author={Zichen Yu and Changyong Shu and Jiajun Deng and Kangjie Lu and Zongdai Liu and Jiangyong Yu and Dawei Yang and Hui Li and Yan Chen},
year={2023},
eprint={2311.12058},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
docker-hub/FlashOCC/Flashocc/data/nuscenes 0 → 120000
View file @ d2b71343
/data/nuScenes
\ No newline at end of file
docker-hub/FlashOCC/Flashocc/doc/cmd.md 0 → 100644
View file @ d2b71343
# Training cmd
## 1. FlashOcc
```shell script
bash tool/dist_train.sh projects/configs/flashocc/flashocc-r50-M0.py 4 # 31.95
bash tool/dist_train.sh projects/configs/flashocc/flashocc-r50.py 4 # 32.08
bash tool/dist_train.sh projects/configs/flashocc/flashocc-r50-4d-stereo.py 4 # 37.84
bash tool/dist_train.sh projects/configs/flashocc/flashocc-stbase-4d-stereo-512x1408_4x4_1e-4.py 4 # 41.80
bash tool/dist_train.sh projects/configs/flashocc/flashocc-stbase-4d-stereo-512x1408_4x4_2e-4.py 4 # 43.52
```
## 2. Panoptic-FlashOcc
### for train
```shell script
conda activate FlashOcc
exp_name=panoptic-flashocc-r50-depth-tiny-pano
exp_name=panoptic-flashocc-r50-depth-pano
exp_name=panoptic-flashocc-r50-depth4d-pano
exp_name=panoptic-flashocc-r50-depth4d-longterm8f-pano
bash tools/dist_train.sh \
projects/configs/panoptic-flashocc/${exp_name}.py \
4
```
### for test
```shell script
conda activate FlashOcc
exp_name=panoptic-flashocc-r50-depth-tiny-pano
exp_name=panoptic-flashocc-r50-depth-pano
exp_name=panoptic-flashocc-r50-depth4d-pano
exp_name=panoptic-flashocc-r50-depth4d-longterm8f-pano
bash tools/dist_test.sh \
projects/configs/panoptic-flashocc/${exp_name}.py \
work_dirs/${exp_name}/epoch_24_ema.pth \
4 \
--eval ray-iou
```
### for vis
```shell script
exp_name=panoptic-flashocc-r50-depth-tiny-pano
exp_name=panoptic-flashocc-r50-depth-pano
exp_name=panoptic-flashocc-r50-depth4d-pano
exp_name=panoptic-flashocc-r50-depth4d-longterm8f-pano
python tools/vis_occ.py --config projects/configs/panoptic-flashocc/${exp_name}.py --weights work_dirs/${exp_name}/epoch_24_ema.pth --viz-dir vis/${exp_name} --draw-gt
```
### for test inference time
```shell script
conda activate FlashOcc
source activate FlashOcc
exp_name=panoptic-flashocc-r50-depth-tiny-pano
exp_name=panoptic-flashocc-r50-depth-pano
python tools/analysis_tools/benchmark.py \
projects/configs/panoptic-flashocc/${exp_name}.py \
work_dirs/${exp_name}/epoch_24_ema.pth \
--w_pano --w_panoproc
exp_name=panoptic-flashocc-r50-depth4d-pano
exp_name=panoptic-flashocc-r50-depth4d-longterm8f-pano
python tools/analysis_tools/benchmark_sequential.py \
projects/configs/panoptic-flashocc/${exp_name}.py \
work_dirs/${exp_name}/epoch_24_ema.pth \
--w_pano --w_panoproc
```
docker-hub/FlashOCC/Flashocc/doc/install.md 0 → 100644
View file @ d2b71343
## Environment Setup
step 1. Install environment for pytorch training
```
conda create --name FlashOcc python=3.8.5
conda activate FlashOcc
pip install torch==1.10.0+cu111 torchvision==0.11.0+cu111 torchaudio==0.10.0 -f https://download.pytorch.org/whl/torch_stable.html
pip install mmcv-full==1.5.3
pip install mmdet==2.25.1
pip install mmsegmentation==0.25.0
sudo apt-get install python3-dev
sudo apt-get install libevent-dev
sudo apt-get groupinstall 'development tools'
export PATH=/usr/local/cuda/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH
export CUDA_ROOT=/usr/local/cuda
pip install pycuda
pip install lyft_dataset_sdk
pip install networkx==2.2
pip install numba==0.53.0
pip install numpy==1.23.5
pip install nuscenes-devkit
pip install plyfile
pip install scikit-image
pip install tensorboard
pip install trimesh==2.35.39
pip install setuptools==59.5.0
pip install yapf==0.40.1
cd Path_to_FlashOcc
git clone git@github.com:Yzichen/FlashOCC.git
cd Path_to_FlashOcc/FlashOcc
git clone https://github.com/open-mmlab/mmdetection3d.git
cd Path_to_FlashOcc/FlashOcc/mmdetection3d
git checkout v1.0.0rc4
pip install -v -e .
cd Path_to_FlashOcc/FlashOcc/projects
pip install -v -e .
```
step 3. Prepare nuScenes dataset as introduced in [nuscenes_det.md](nuscenes_det.md) and create the pkl for FlashOCC by running:
```shell
python tools/create_data_bevdet.py
```
thus, the folder will be ranged as following:
```shell script
└── Path_to_FlashOcc/
└── data
└── nuscenes
├── v1.0-trainval (existing)
├── sweeps (existing)
├── samples (existing)
├── bevdetv2-nuscenes_infos_train.pkl (new)
└── bevdetv2-nuscenes_infos_val.pkl (new)
```
step 4. For Occupancy Prediction task, download (only) the 'gts' from [CVPR2023-3D-Occupancy-Prediction](https://github.com/CVPR2023-3D-Occupancy-Prediction/CVPR2023-3D-Occupancy-Prediction) and arrange the folder as:
```shell script
└── Path_to_FlashOcc/
└── data
└── nuscenes
├── v1.0-trainval (existing)
├── sweeps (existing)
├── samples (existing)
├── gts (new)
├── bevdetv2-nuscenes_infos_train.pkl (new)
└── bevdetv2-nuscenes_infos_val.pkl (new)
```
(for panoptic occupancy), we follow the data setting in SparseOcc:
(1) Download Occ3D-nuScenes occupancy GT from [gdrive](https://drive.google.com/file/d/1kiXVNSEi3UrNERPMz_CfiJXKkgts_5dY/view?usp=drive_link), unzip it, and save it to `data/nuscenes/occ3d`.
(2) Generate the panoptic occupancy ground truth with `gen_instance_info.py`. The panoptic version of Occ3D will be saved to `data/nuscenes/occ3d_panoptic`.
step 5. CKPTS Preparation
(1) Download flashocc-r50-256x704.pth[https://drive.google.com/file/d/1k9BzXB2nRyvXhqf7GQx3XNSej6Oq6I-B/view] to Path_to_FlashOcc/FlashOcc/ckpts/, then run:
```shell script
bash tools/dist_test.sh projects/configs/flashocc/flashocc-r50.py ckpts/flashocc-r50-256x704.pth 4 --eval map
```
step 6. (Optional) Install mmdeploy for tensorrt testing
```shell script
conda activate FlashOcc
pip install Cython==0.29.24
### get tensorrt
wget https://developer.download.nvidia.com/compute/machine-learning/tensorrt/secure/8.4.0/tars/TensorRT-8.4.0.6.Linux.x86_64-gnu.cuda-11.6.cudnn8.3.tar.gz
export TENSORRT_DIR=Path_to_TensorRT-8.4.0.6
### get onnxruntime
ONNXRUNTIME_VERSION=1.8.1
pip install onnxruntime-gpu==${ONNXRUNTIME_VERSION}
cd Path_to_your_onnxruntime
wget https://github.com/microsoft/onnxruntime/releases/download/v${ONNXRUNTIME_VERSION}/onnxruntime-linux-x64-${ONNXRUNTIME_VERSION}.tgz \
&& tar -zxvf onnxruntime-linux-x64-${ONNXRUNTIME_VERSION}.tgz
# export ONNXRUNTIME_DIR=/data01/shuchangyong/pkgs/onnxruntime-linux-x64-1.8.1
export ONNXRUNTIME_DIR=Path_to_your_onnxruntime/onnxruntime-linux-x64-1.8.1
cd Path_to_FlashOcc/FlashOcc/
git clone git@github.com:drilistbox/mmdeploy.git
cd Path_to_FlashOcc/FlashOcc/mmdeploy
git submodule update --init --recursive
mkdir -p build
cd Path_to_FlashOcc/FlashOcc/mmdeploy/build
cmake -DMMDEPLOY_TARGET_BACKENDS="ort;trt" ..
make -j 16
cd Path_to_FlashOcc/FlashOcc/mmdeploy
pip install -e .
### build sdk
cd Path_to_pplcv/
git clone https://github.com/openppl-public/ppl.cv.git
cd Path_to_pplcv/ppl.cv
export PPLCV_VERSION=0.7.0
git checkout tags/v${PPLCV_VERSION} -b v${PPLCV_VERSION}
./build.sh cuda
#pip install nvidia-tensorrt==8.4.0.6
pip install nvidia-tensorrt==8.4.1.5
pip install tensorrt
#pip install h5py
pip install spconv==2.3.6
export PATH=Path_to_TensorRT-8.4.0.6/bin:$PATH
export LD_LIBRARY_PATH=Path_to_TensorRT-8.4.0.6/lib:$LD_LIBRARY_PATH
export LIBRARY_PATH=Path_to_TensorRT-8.4.0.6/lib:$LIBRARY_PATH
```
## The finally overall rangement
1. Tensort
```shell script
└── Path_to_TensorRT-8.4.0.6
└── TensorRT-8.4.0.6
```
2. FlashOcc
```shell script
└── Path_to_FlashOcc/
└── data
└── nuscenes
├── v1.0-trainval (existing)
├── sweeps (existing)
├── samples (existing)
├── gts (new)
├── bevdetv2-nuscenes_infos_train.pkl (new)
└── bevdetv2-nuscenes_infos_val.pkl (new)
└── doc
├── install.md
└── trt_test.md
├── figs
├── mmdeploy (new)
├── mmdetection3d (new)
├── projects
├── requirements
├── tools
└── README.md
```
3. ppl.cv
```shell script
└── Path_to_pplcv
└── ppl.cv
```
docker-hub/FlashOCC/Flashocc/doc/mmdeploy_test.md 0 → 100644
View file @ d2b71343
# trt inference speed
```shell
conda activate FlashOcc
1. cmd for M0
exp_name=flashocc-r50-M0
fold_name=flashocc
config=projects/configs/${fold_name}/${exp_name}-trt.py
checkpoint=ckpts/flashocc-r50-M0-256x704.pth
work_dir=work_dirs/${exp_name}/onnx_trt/
2. cmd for M1
exp_name=flashocc-r50
fold_name=flashocc
config=projects/configs/${fold_name}/${exp_name}-trt.py
checkpoint=ckpts/flashocc-r50-256x704.pth
work_dir=work_dirs/${exp_name}/onnx_trt/
# int8 test.
engine=work_dirs/${exp_name}/onnx_trt/bevdet_int8_fuse.engine
python tools/convert_bevdet_to_TRT.py $config $checkpoint $work_dir --fuse-conv-bn --int8 --calib_num 256
python tools/analysis_tools/benchmark_trt.py $config $engine --eval
# ===> per class IoU of 6019 samples:
# ===> others - IoU = 4.58
# ===> barrier - IoU = 34.13
# ===> bicycle - IoU = 8.68
# ===> bus - IoU = 34.9
# ===> car - IoU = 40.48
# ===> construction_vehicle - IoU = 15.99
# ===> motorcycle - IoU = 15.49
# ===> pedestrian - IoU = 13.58
# ===> traffic_cone - IoU = 12.83
# ===> trailer - IoU = 25.31
# ===> truck - IoU = 28.08
# ===> driveable_surface - IoU = 76.7
# ===> other_flat - IoU = 31.5
# ===> sidewalk - IoU = 45.01
# ===> terrain - IoU = 49.63
# ===> manmade - IoU = 35.72
# ===> vegetation - IoU = 30.39
# ===> mIoU of 6019 samples: 29.59
# int8+fp16 test.
engine=work_dirs/${exp_name}/onnx_trt/bevdet_int8_fp16_fuse.engine
python tools/convert_bevdet_to_TRT.py $config $checkpoint $work_dir --fuse-conv-bn --fp16 --int8 --calib_num 256
python tools/analysis_tools/benchmark_trt.py $config $engine --eval
# ===> per class IoU of 6019 samples:
# ===> others - IoU = 4.59
# ===> barrier - IoU = 34.13
# ===> bicycle - IoU = 8.71
# ===> bus - IoU = 34.9
# ===> car - IoU = 40.49
# ===> construction_vehicle - IoU = 16.01
# ===> motorcycle - IoU = 15.55
# ===> pedestrian - IoU = 13.63
# ===> traffic_cone - IoU = 12.86
# ===> trailer - IoU = 25.33
# ===> truck - IoU = 28.1
# ===> driveable_surface - IoU = 76.7
# ===> other_flat - IoU = 31.51
# ===> sidewalk - IoU = 45.01
# ===> terrain - IoU = 49.63
# ===> manmade - IoU = 35.72
# ===> vegetation - IoU = 30.39
# ===> mIoU of 6019 samples: 29.6
# fp16 test
engine=work_dirs/${exp_name}/onnx_trt/bevdet_fp16_fuse.engine
python tools/convert_bevdet_to_TRT.py $config $checkpoint $work_dir --fuse-conv-bn --fp16
python tools/analysis_tools/benchmark_trt.py $config $engine
python tools/analysis_tools/benchmark_trt.py $config $engine --eval
# ===> per class IoU of 6019 samples:
# ===> others - IoU = 5.97
# ===> barrier - IoU = 36.37
# ===> bicycle - IoU = 10.14
# ===> bus - IoU = 35.47
# ===> car - IoU = 41.57
# ===> construction_vehicle - IoU = 15.73
# ===> motorcycle - IoU = 14.8
# ===> pedestrian - IoU = 15.65
# ===> traffic_cone - IoU = 14.46
# ===> trailer - IoU = 27.47
# ===> truck - IoU = 29.39
# ===> driveable_surface - IoU = 77.14
# ===> other_flat - IoU = 34.66
# ===> sidewalk - IoU = 46.44
# ===> terrain - IoU = 51.05
# ===> manmade - IoU = 35.79
# ===> vegetation - IoU = 31.19
# ===> mIoU of 6019 samples: 30.78
# fp32 test
engine=work_dirs/${exp_name}/onnx_trt/bevdet_fuse.engine
python tools/convert_bevdet_to_TRT.py $config $checkpoint $work_dir --fuse-conv-bn
python tools/analysis_tools/benchmark_trt.py $config $engine
python tools/analysis_tools/benchmark_trt.py $config $engine --eval
# ===> per class IoU of 6019 samples:
# ===> others - IoU = 5.97
# ===> barrier - IoU = 36.37
# ===> bicycle - IoU = 10.15
# ===> bus - IoU = 35.46
# ===> car - IoU = 41.56
# ===> construction_vehicle - IoU = 15.73
# ===> motorcycle - IoU = 14.78
# ===> pedestrian - IoU = 15.64
# ===> traffic_cone - IoU = 14.44
# ===> trailer - IoU = 27.46
# ===> truck - IoU = 29.39
# ===> driveable_surface - IoU = 77.14
# ===> other_flat - IoU = 34.68
# ===> sidewalk - IoU = 46.44
# ===> terrain - IoU = 51.05
# ===> manmade - IoU = 35.79
# ===> vegetation - IoU = 31.18
# ===> mIoU of 6019 samples: 30.78
```
3. cmd for flashoccv2
```
exp_name=flashoccv2-r50-depth
fold_name=flashoccv2
config=projects/configs/${fold_name}/${exp_name}-trt.py
checkpoint=work_dirs/${exp_name}/epoch_24_ema.pth
work_dir=work_dirs/${exp_name}/onnx_trt/
# fp16 test
engine=work_dirs/${exp_name}/onnx_trt/bevdet_fp16_fuse.engine
python tools/convert_bevdet_to_TRT.py $config $checkpoint $work_dir --fuse-conv-bn --fp16
python tools/analysis_tools/benchmark_trt.py $config $engine
python tools/analysis_tools/benchmark_trt.py $config $engine --eval
```
# Flops and params
```shell
python tools/analysis_tools/get_flops.py projects/configs/bevdet_occ/bevdet-occ-r50.py --modality image --shape 256 704
python tools/analysis_tools/get_flops.py projects/configs/flashocc/flashocc-r50-M0.py --modality image --shape 256 704
python tools/analysis_tools/get_flops.py projects/configs/flashocc/flashocc-r50.py --modality image --shape 256 704
python tools/analysis_tools/get_flops.py projects/configs/flashocc/flashocc-stbase-4d-stereo-512x1408.py --modality image --shape 512 1408
python tools/analysis_tools/get_flops.py projects/configs/flashoccv2/flashoccv2-r50-depth.py --modality image --shape 256 704
python tools/analysis_tools/get_flops.py projects/configs/flashoccv2/flashoccv2-r50-depth-tiny.py --modality image --shape 256 704
```
\ No newline at end of file
docker-hub/FlashOCC/Flashocc/doc/model_training.md 0 → 100644
View file @ d2b71343
#### Train model
```shell
# single gpu
python tools/train.py $config
# multiple gpu
./tools/dist_train.sh $config num_gpu
```
#### Test model
```shell
# single gpu
python tools/test.py $config $checkpoint --eval mAP
# multiple gpu
./tools/dist_test.sh $config $checkpoint num_gpu --eval mAP
# ray-iou metric
./tools/dist_test.sh $config $checkpoint num_gpu --eval ray-iou
```
#### FPS for Panoptic-FlashOcc
```shell
# for single-frame
python tools/analysis_tools/benchmark.py config ckpt
python tools/analysis_tools/benchmark.py config ckpt --w_pano
# for multi-frame
python tools/analysis_tools/benchmark_sequential.py config ckpt
python tools/analysis_tools/benchmark_sequential.py config ckpt --w_pano
```
docker-hub/FlashOCC/Flashocc/doc/nuscenes_det.md 0 → 100644
View file @ d2b71343
# NuScenes Dataset for 3D Object Detection
This page provides specific tutorials about the usage of MMDetection3D for nuScenes dataset.
## Before Preparation
You can download nuScenes 3D detection data [HERE](https://www.nuscenes.org/download) and unzip all zip files.
Like the general way to prepare dataset, it is recommended to symlink the dataset root to `$MMDETECTION3D/data`.
The folder structure should be organized as follows before our processing.
```
mmdetection3d
├── mmdet3d
├── tools
├── configs
├── data
│ ├── nuscenes
│ │ ├── maps
│ │ ├── samples
│ │ ├── sweeps
│ │ ├── v1.0-test
| | ├── v1.0-trainval
```
## Dataset Preparation
We typically need to organize the useful data information with a .pkl or .json file in a specific style, e.g., coco-style for organizing images and their annotations.
To prepare these files for nuScenes, run the following command:
```bash
python tools/create_data.py nuscenes --root-path ./data/nuscenes --out-dir ./data/nuscenes --extra-tag nuscenes
```
The folder structure after processing should be as below.
```
mmdetection3d
├── mmdet3d
├── tools
├── configs
├── data
│ ├── nuscenes
│ │ ├── maps
│ │ ├── samples
│ │ ├── sweeps
│ │ ├── v1.0-test
| | ├── v1.0-trainval
│ │ ├── nuscenes_database
│ │ ├── nuscenes_infos_train.pkl
│ │ ├── nuscenes_infos_val.pkl
│ │ ├── nuscenes_infos_test.pkl
│ │ ├── nuscenes_dbinfos_train.pkl
│ │ ├── nuscenes_infos_train_mono3d.coco.json
│ │ ├── nuscenes_infos_val_mono3d.coco.json
│ │ ├── nuscenes_infos_test_mono3d.coco.json
```
Here, .pkl files are generally used for methods involving point clouds and coco-style .json files are more suitable for image-based methods, such as image-based 2D and 3D detection.
Next, we will elaborate on the details recorded in these info files.
- `nuscenes_database/xxxxx.bin`: point cloud data included in each 3D bounding box of the training dataset
- `nuscenes_infos_train.pkl`: training dataset info, each frame info has two keys: `metadata` and `infos`.
`metadata` contains the basic information for the dataset itself, such as `{'version': 'v1.0-trainval'}`, while `infos` contains the detailed information as follows:
- info\['lidar_path'\]: The file path of the lidar point cloud data.
- info\['token'\]: Sample data token.
- info\['sweeps'\]: Sweeps information (`sweeps` in the nuScenes refer to the intermediate frames without annotations, while `samples` refer to those key frames with annotations).
- info\['sweeps'\]\[i\]\['data_path'\]: The data path of i-th sweep.
- info\['sweeps'\]\[i\]\['type'\]: The sweep data type, e.g., `'lidar'`.
- info\['sweeps'\]\[i\]\['sample_data_token'\]: The sweep sample data token.
- info\['sweeps'\]\[i\]\['sensor2ego_translation'\]: The translation from the current sensor (for collecting the sweep data) to ego vehicle. (1x3 list)
- info\['sweeps'\]\[i\]\['sensor2ego_rotation'\]: The rotation from the current sensor (for collecting the sweep data) to ego vehicle. (1x4 list in the quaternion format)
- info\['sweeps'\]\[i\]\['ego2global_translation'\]: The translation from the ego vehicle to global coordinates. (1x3 list)
- info\['sweeps'\]\[i\]\['ego2global_rotation'\]: The rotation from the ego vehicle to global coordinates. (1x4 list in the quaternion format)
- info\['sweeps'\]\[i\]\['timestamp'\]: Timestamp of the sweep data.
- info\['sweeps'\]\[i\]\['sensor2lidar_translation'\]: The translation from the current sensor (for collecting the sweep data) to lidar. (1x3 list)
- info\['sweeps'\]\[i\]\['sensor2lidar_rotation'\]: The rotation from the current sensor (for collecting the sweep data) to lidar. (1x4 list in the quaternion format)
- info\['cams'\]: Cameras calibration information. It contains six keys corresponding to each camera: `'CAM_FRONT'`, `'CAM_FRONT_RIGHT'`, `'CAM_FRONT_LEFT'`, `'CAM_BACK'`, `'CAM_BACK_LEFT'`, `'CAM_BACK_RIGHT'`.
Each dictionary contains detailed information following the above way for each sweep data (has the same keys for each information as above). In addition, each camera has a key `'cam_intrinsic'` for recording the intrinsic parameters when projecting 3D points to each image plane.
- info\['lidar2ego_translation'\]: The translation from lidar to ego vehicle. (1x3 list)
- info\['lidar2ego_rotation'\]: The rotation from lidar to ego vehicle. (1x4 list in the quaternion format)
- info\['ego2global_translation'\]: The translation from the ego vehicle to global coordinates. (1x3 list)
- info\['ego2global_rotation'\]: The rotation from the ego vehicle to global coordinates. (1x4 list in the quaternion format)
- info\['timestamp'\]: Timestamp of the sample data.
- info\['gt_boxes'\]: 7-DoF annotations of 3D bounding boxes, an Nx7 array.
- info\['gt_names'\]: Categories of 3D bounding boxes, an 1xN array.
- info\['gt_velocity'\]: Velocities of 3D bounding boxes (no vertical measurements due to inaccuracy), an Nx2 array.
- info\['num_lidar_pts'\]: Number of lidar points included in each 3D bounding box.
- info\['num_radar_pts'\]: Number of radar points included in each 3D bounding box.
- info\['valid_flag'\]: Whether each bounding box is valid. In general, we only take the 3D boxes that include at least one lidar or radar point as valid boxes.
- `nuscenes_infos_train_mono3d.coco.json`: training dataset coco-style info. This file organizes image-based data into three categories (keys): `'categories'`, `'images'`, `'annotations'`.
- info\['categories'\]: A list containing all the category names. Each element follows the dictionary format and consists of two keys: `'id'` and `'name'`.
- info\['images'\]: A list containing all the image info.
- info\['images'\]\[i\]\['file_name'\]: The file name of the i-th image.
- info\['images'\]\[i\]\['id'\]: Sample data token of the i-th image.
- info\['images'\]\[i\]\['token'\]: Sample token corresponding to this frame.
- info\['images'\]\[i\]\['cam2ego_rotation'\]: The rotation from the camera to ego vehicle. (1x4 list in the quaternion format)
- info\['images'\]\[i\]\['cam2ego_translation'\]: The translation from the camera to ego vehicle. (1x3 list)
- info\['images'\]\[i\]\['ego2global_rotation''\]: The rotation from the ego vehicle to global coordinates. (1x4 list in the quaternion format)
- info\['images'\]\[i\]\['ego2global_translation'\]: The translation from the ego vehicle to global coordinates. (1x3 list)
- info\['images'\]\[i\]\['cam_intrinsic'\]: Camera intrinsic matrix. (3x3 list)
- info\['images'\]\[i\]\['width'\]: Image width, 1600 by default in nuScenes.
- info\['images'\]\[i\]\['height'\]: Image height, 900 by default in nuScenes.
- info\['annotations'\]: A list containing all the annotation info.
- info\['annotations'\]\[i\]\['file_name'\]: The file name of the corresponding image.
- info\['annotations'\]\[i\]\['image_id'\]: The image id (token) of the corresponding image.
- info\['annotations'\]\[i\]\['area'\]: Area of the 2D bounding box.
- info\['annotations'\]\[i\]\['category_name'\]: Category name.
- info\['annotations'\]\[i\]\['category_id'\]: Category id.
- info\['annotations'\]\[i\]\['bbox'\]: 2D bounding box annotation (exterior rectangle of the projected 3D box), 1x4 list following \[x1, y1, x2-x1, y2-y1\].
x1/y1 are minimum coordinates along horizontal/vertical direction of the image.
- info\['annotations'\]\[i\]\['iscrowd'\]: Whether the region is crowded. Defaults to 0.
- info\['annotations'\]\[i\]\['bbox_cam3d'\]: 3D bounding box (gravity) center location (3), size (3), (global) yaw angle (1), 1x7 list.
- info\['annotations'\]\[i\]\['velo_cam3d'\]: Velocities of 3D bounding boxes (no vertical measurements due to inaccuracy), an Nx2 array.
- info\['annotations'\]\[i\]\['center2d'\]: Projected 3D-center containing 2.5D information: projected center location on the image (2) and depth (1), 1x3 list.
- info\['annotations'\]\[i\]\['attribute_name'\]: Attribute name.
- info\['annotations'\]\[i\]\['attribute_id'\]: Attribute id.
We maintain a default attribute collection and mapping for attribute classification.
Please refer to [here](https://github.com/open-mmlab/mmdetection3d/blob/master/mmdet3d/datasets/nuscenes_mono_dataset.py#L53) for more details.
- info\['annotations'\]\[i\]\['id'\]: Annotation id. Defaults to `i`.
Here we only explain the data recorded in the training info files. The same applies to validation and testing set.
The core function to get `nuscenes_infos_xxx.pkl` and `nuscenes_infos_xxx_mono3d.coco.json` are [\_fill_trainval_infos](https://github.com/open-mmlab/mmdetection3d/blob/master/tools/data_converter/nuscenes_converter.py#L143) and [get_2d_boxes](https://github.com/open-mmlab/mmdetection3d/blob/master/tools/data_converter/nuscenes_converter.py#L397), respectively.
Please refer to [nuscenes_converter.py](https://github.com/open-mmlab/mmdetection3d/blob/master/tools/data_converter/nuscenes_converter.py) for more details.
## Training pipeline
### LiDAR-Based Methods
A typical training pipeline of LiDAR-based 3D detection (including multi-modality methods) on nuScenes is as below.
```python
train_pipeline = [
dict(
type='LoadPointsFromFile',
coord_type='LIDAR',
load_dim=5,
use_dim=5,
file_client_args=file_client_args),
dict(
type='LoadPointsFromMultiSweeps',
sweeps_num=10,
file_client_args=file_client_args),
dict(type='LoadAnnotations3D', with_bbox_3d=True, with_label_3d=True),
dict(
type='GlobalRotScaleTrans',
rot_range=[-0.3925, 0.3925],
scale_ratio_range=[0.95, 1.05],
translation_std=[0, 0, 0]),
dict(type='RandomFlip3D', flip_ratio_bev_horizontal=0.5),
dict(type='PointsRangeFilter', point_cloud_range=point_cloud_range),
dict(type='ObjectRangeFilter', point_cloud_range=point_cloud_range),
dict(type='ObjectNameFilter', classes=class_names),
dict(type='PointShuffle'),
dict(type='DefaultFormatBundle3D', class_names=class_names),
dict(type='Collect3D', keys=['points', 'gt_bboxes_3d', 'gt_labels_3d'])
]
```
Compared to general cases, nuScenes has a specific `'LoadPointsFromMultiSweeps'` pipeline to load point clouds from consecutive frames. This is a common practice used in this setting.
Please refer to the nuScenes [original paper](https://arxiv.org/abs/1903.11027) for more details.
The default `use_dim` in `'LoadPointsFromMultiSweeps'` is `[0, 1, 2, 4]`, where the first 3 dimensions refer to point coordinates and the last refers to timestamp differences.
Intensity is not used by default due to its yielded noise when concatenating the points from different frames.
### Vision-Based Methods
A typical training pipeline of image-based 3D detection on nuScenes is as below.
```python
train_pipeline = [
dict(type='LoadImageFromFileMono3D'),
dict(
type='LoadAnnotations3D',
with_bbox=True,
with_label=True,
with_attr_label=True,
with_bbox_3d=True,
with_label_3d=True,
with_bbox_depth=True),
dict(type='Resize', img_scale=(1600, 900), keep_ratio=True),
dict(type='RandomFlip3D', flip_ratio_bev_horizontal=0.5),
dict(type='Normalize', **img_norm_cfg),
dict(type='Pad', size_divisor=32),
dict(type='DefaultFormatBundle3D', class_names=class_names),
dict(
type='Collect3D',
keys=[
'img', 'gt_bboxes', 'gt_labels', 'attr_labels', 'gt_bboxes_3d',
'gt_labels_3d', 'centers2d', 'depths'
]),
]
```
It follows the general pipeline of 2D detection while differs in some details:
- It uses monocular pipelines to load images, which includes additional required information like camera intrinsics.
- It needs to load 3D annotations.
- Some data augmentation techniques need to be adjusted, such as `RandomFlip3D`.
Currently we do not support more augmentation methods, because how to transfer and apply other techniques is still under explored.
## Evaluation
An example to evaluate PointPillars with 8 GPUs with nuScenes metrics is as follows.
```shell
bash ./tools/dist_test.sh configs/pointpillars/hv_pointpillars_fpn_sbn-all_4x8_2x_nus-3d.py checkpoints/hv_pointpillars_fpn_sbn-all_4x8_2x_nus-3d_20200620_230405-2fa62f3d.pth 8 --eval bbox
```
## Metrics
NuScenes proposes a comprehensive metric, namely nuScenes detection score (NDS), to evaluate different methods and set up the benchmark.
It consists of mean Average Precision (mAP), Average Translation Error (ATE), Average Scale Error (ASE), Average Orientation Error (AOE), Average Velocity Error (AVE) and Average Attribute Error (AAE).
Please refer to its [official website](https://www.nuscenes.org/object-detection?externalData=all&mapData=all&modalities=Any) for more details.
We also adopt this approach for evaluation on nuScenes. An example of printed evaluation results is as follows:
```
mAP: 0.3197
mATE: 0.7595
mASE: 0.2700
mAOE: 0.4918
mAVE: 1.3307
mAAE: 0.1724
NDS: 0.3905
Eval time: 170.8s
Per-class results:
Object Class AP ATE ASE AOE AVE AAE
car 0.503 0.577 0.152 0.111 2.096 0.136
truck 0.223 0.857 0.224 0.220 1.389 0.179
bus 0.294 0.855 0.204 0.190 2.689 0.283
trailer 0.081 1.094 0.243 0.553 0.742 0.167
construction_vehicle 0.058 1.017 0.450 1.019 0.137 0.341
pedestrian 0.392 0.687 0.284 0.694 0.876 0.158
motorcycle 0.317 0.737 0.265 0.580 2.033 0.104
bicycle 0.308 0.704 0.299 0.892 0.683 0.010
traffic_cone 0.555 0.486 0.309 nan nan nan
barrier 0.466 0.581 0.269 0.169 nan nan
```
## Testing and make a submission
An example to test PointPillars on nuScenes with 8 GPUs and generate a submission to the leaderboard is as follows.
```shell
./tools/dist_test.sh configs/pointpillars/hv_pointpillars_fpn_sbn-all_4x8_2x_nus-3d.py work_dirs/pp-nus/latest.pth 8 --out work_dirs/pp-nus/results_eval.pkl --format-only --eval-options 'jsonfile_prefix=work_dirs/pp-nus/results_eval'
```
Note that the testing info should be changed to that for testing set instead of validation set [here](https://github.com/open-mmlab/mmdetection3d/blob/master/configs/_base_/datasets/nus-3d.py#L132).
After generating the `work_dirs/pp-nus/results_eval.json`, you can compress it and submit it to nuScenes benchmark. Please refer to the [nuScenes official website](https://www.nuscenes.org/object-detection?externalData=all&mapData=all&modalities=Any) for more information.
We can also visualize the prediction results with our developed visualization tools. Please refer to the [visualization doc](https://mmdetection3d.readthedocs.io/en/latest/useful_tools.html#visualization) for more details.
## Notes
### Transformation between `NuScenesBox` and our `CameraInstanceBoxes`.
In general, the main difference of `NuScenesBox` and our `CameraInstanceBoxes` is mainly reflected in the yaw definition. `NuScenesBox` defines the rotation with a quaternion or three Euler angles while ours only defines one yaw angle due to the practical scenario. It requires us to add some additional rotations manually in the pre-processing and post-processing, such as [here](https://github.com/open-mmlab/mmdetection3d/blob/master/mmdet3d/datasets/nuscenes_mono_dataset.py#L673).
In addition, please note that the definition of corners and locations are detached in the `NuScenesBox`. For example, in monocular 3D detection, the definition of the box location is in its camera coordinate (see its official [illustration](https://www.nuscenes.org/nuscenes#data-collection) for car setup), which is consistent with [ours](https://github.com/open-mmlab/mmdetection3d/blob/master/mmdet3d/core/bbox/structures/cam_box3d.py). In contrast, its corners are defined with the [convention](https://github.com/nutonomy/nuscenes-devkit/blob/02e9200218977193a1058dd7234f935834378319/python-sdk/nuscenes/utils/data_classes.py#L527) "x points forward, y to the left, z up". It results in different philosophy of dimension and rotation definitions from our `CameraInstanceBoxes`. An example to remove similar hacks is PR [#744](https://github.com/open-mmlab/mmdetection3d/pull/744). The same problem also exists in the LiDAR system. To deal with them, we typically add some transformation in the pre-processing and post-processing to guarantee the box will be in our coordinate system during the entire training and inference procedure.
docker-hub/FlashOCC/Flashocc/doc/visualization.md 0 → 100644
View file @ d2b71343
# FlashOcc
```shell
# step 1. generate result
bash tools/dist_test.sh projects/configs/flashocc/flashocc-r50.py ckpts/flashocc-r50-256x704.pth 4 --eval map --eval-options show_dir=work_dirs/flashocc_r50/results
# step 2. visualization
python tools/analysis_tools/vis_occ.py work_dirs/flashocc_r50/results/ --root_path ./data/nuscenes --save_path ./vis
```
# Panoptic-FlashOcc
```shell
exp_name=panoptic-flashocc-r50-depth4d-longterm8f-pano
python tools/vis_occ.py --config projects/configs/panoptic-flashocc/${exp_name}.py --weights work_dirs/${exp_name}/epoch_24_ema.pth --viz-dir vis/${exp_name} --draw-pano-gt
```
docker-hub/FlashOCC/Flashocc/figs/visualization.png 0 → 100644
View file @ d2b71343
This image diff could not be displayed because it is too large. You can view the blob instead.
docker-hub/FlashOCC/Flashocc/lib/dvr/dvr.cpp 0 → 100644
View file @ d2b71343
// Acknowledgments: https://github.com/tarashakhurana/4d-occ-forecasting
// Modified by Haisong Liu
#include <string>
#include <torch/extension.h>
#include <vector>
/*
* CUDA forward declarations
*/
std::vector<torch::Tensor> render_forward_cuda(torch::Tensor sigma,
torch::Tensor origin,
torch::Tensor points,
torch::Tensor tindex,
const std::vector<int> grid,
std::string phase_name);
std::vector<torch::Tensor>
render_cuda(torch::Tensor sigma, torch::Tensor origin, torch::Tensor points,
torch::Tensor tindex, std::string loss_name);
torch::Tensor init_cuda(torch::Tensor points, torch::Tensor tindex,
const std::vector<int> grid);
/*
* C++ interface
*/
#define CHECK_CUDA(x) \
TORCH_CHECK(x.type().is_cuda(), #x " must be a CUDA tensor")
#define CHECK_CONTIGUOUS(x) \
TORCH_CHECK(x.is_contiguous(), #x " must be contiguous")
#define CHECK_INPUT(x) \
CHECK_CUDA(x); \
CHECK_CONTIGUOUS(x)
std::vector<torch::Tensor>
render_forward(torch::Tensor sigma, torch::Tensor origin, torch::Tensor points,
torch::Tensor tindex, const std::vector<int> grid,
std::string phase_name) {
CHECK_INPUT(sigma);
CHECK_INPUT(origin);
CHECK_INPUT(points);
CHECK_INPUT(tindex);
return render_forward_cuda(sigma, origin, points, tindex, grid, phase_name);
}
std::vector<torch::Tensor> render(torch::Tensor sigma, torch::Tensor origin,
torch::Tensor points, torch::Tensor tindex,
std::string loss_name) {
CHECK_INPUT(sigma);
CHECK_INPUT(origin);
CHECK_INPUT(points);
CHECK_INPUT(tindex);
return render_cuda(sigma, origin, points, tindex, loss_name);
}
torch::Tensor init(torch::Tensor points, torch::Tensor tindex,
const std::vector<int> grid) {
CHECK_INPUT(points);
CHECK_INPUT(tindex);
return init_cuda(points, tindex, grid);
}
PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) {
m.def("init", &init, "Initialize");
m.def("render", &render, "Render");
m.def("render_forward", &render_forward, "Render (forward pass only)");
}
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment