文件列表:

.pre-commit-config.yaml (1426, 2023-12-07)
.readthedocs.yml (196, 2023-12-07)
CITATION.cff (304, 2023-12-07)
LICENSE (4164, 2023-12-07)
MANIFEST.in (216, 2023-12-07)
docs/ (0, 2023-12-07)
docs/en/ (0, 2023-12-07)
docs/en/1_exist_data_model.md (14707, 2023-12-07)
docs/en/2_new_data_model.md (6812, 2023-12-07)
docs/en/Makefile (634, 2023-12-07)
docs/en/_static/ (0, 2023-12-07)
docs/en/_static/css/ (0, 2023-12-07)
docs/en/_static/css/readthedocs.css (146, 2023-12-07)
docs/en/api.rst (1319, 2023-12-07)
docs/en/benchmarks.md (12945, 2023-12-07)
docs/en/changelog.md (45381, 2023-12-07)
docs/en/compatibility.md (18080, 2023-12-07)
docs/en/conf.py (4956, 2023-12-07)
docs/en/data_preparation.md (6585, 2023-12-07)
docs/en/datasets/ (0, 2023-12-07)
docs/en/datasets/index.rst (174, 2023-12-07)
docs/en/datasets/kitti_det.md (9410, 2023-12-07)
docs/en/datasets/lyft_det.md (10742, 2023-12-07)
docs/en/datasets/nuscenes_det.md (16856, 2023-12-07)
docs/en/datasets/s3dis_sem_seg.md (11137, 2023-12-07)
docs/en/datasets/scannet_det.md (15726, 2023-12-07)
docs/en/datasets/scannet_sem_seg.md (6122, 2023-12-07)
docs/en/datasets/sunrgbd_det.md (13796, 2023-12-07)
docs/en/datasets/waymo_det.md (10220, 2023-12-07)
docs/en/demo.md (5380, 2023-12-07)
docs/en/faq.md (5566, 2023-12-07)
docs/en/getting_started.md (10713, 2023-12-07)
docs/en/index.rst (1310, 2023-12-07)
docs/en/make.bat (760, 2023-12-07)
docs/en/model_zoo.md (5301, 2023-12-07)
docs/en/news.md (2285, 2023-12-07)
docs/en/stat.py (1565, 2023-12-07)
docs/en/supported_tasks/ (0, 2023-12-07)
... ...

# Forward-Backward View Transformation for Vision-Centric AV Perception

### [Paper (FB-BEV)](https://arxiv.org/abs/2308.02236) | [Paper (FB-OCC)](https://arxiv.org/abs/2307.01492) | [Intro Video](https://youtu.be/KEn8oklzyvo?si=zydZjIwoTzTd-bsO) FB-BEV and FB-OCC are a family of vision-centric 3D object detection and occupancy prediction methods based on forward-backward view transformation. ## News - `[2023/8/01]` FB-BEV was accepted to ICCV 2023. - `[2023/6/16]` FB-OCC wins both Outstanding Champion and Innovation Award in [Autonomous Driving Challenge](https://opendrivelab.com/AD23Challenge.html#Track3) in conjunction with CVPR 2023 End-to-End Autonomous Driving Workshop and Vision-Centric Autonomous Driving Workshop. ## Getting Started - [Installation](docs/install.md) - [Prepare Dataset](docs/prepare_datasets.md) - [Training, Eval, Visualization](docs/start.md) ## Model Zoo | Backbone | Method | Lr Schd | IoU| Config | Download | | :---: | :---: | :---: | :---: | :---: | :---: | | R50 | FB-OCC | 20ep | 39.1 |[config](occupancy_configs/fb_occ/fbocc-r50-cbgs_depth_16f_16x4_20e.py) |[model](https://github.com/zhiqi-li/storage/releases/download/v1.0/fbocc-r50-cbgs_depth_16f_16x4_20e.pth)| * More model weights will be released later. ## Acknowledgement Many thanks to these excellent open source projects: - [BEVFormer](https://github.com/fundamentalvision/BEVFormer), [BEVDet](https://github.com/HuangJunJie2017/BEVDet), [Occ3D](https://github.com/Tsinghua-MARS-Lab/Occ3D), [OpenOccupancy](https://github.com/JeffWang987/OpenOccupancy), [SoloFusion](https://github.com/Divadi/SOLOFusion) ## BibTeX If this work is helpful for your research, please consider citing: ``` @inproceedings{li2023fbbev, title={{FB-BEV}: {BEV} Representation from Forward-Backward View Transformations}, author={Li, Zhiqi and Yu, Zhiding and Wang, Wenhai and Anandkumar, Anima and Lu, Tong and Alvarez, Jose M}, booktitle={IEEE/CVF International Conference on Computer Vision (ICCV)}, year={2023} } ``` ``` @article{li2023fbocc, title={{FB-OCC}: {3D} Occupancy Prediction based on Forward-Backward View Transformation}, author={Li, Zhiqi and Yu, Zhiding and Austin, David and Fang, Mingsheng and Lan, Shiyi and Kautz, Jan and Alvarez, Jose M}, journal={arXiv:2307.01492}, year={2023} } ```

近期下载者：

相关文件：

评论：[我要评论] [举报此文件]

收藏者：