Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
[ECCV 2022] This is the official implementation of BEVFormer, a camera-only framework for autonomous driving perception, e.g., 3D object detection and semantic map segmentation.
| Date | Stars |
|---|---|
| 2026-07-30 | 4558 |
| 2026-08-10 | 4566 |
| 2026-08-18 | 4574 |
| 2026-08-19 | 4574 |
| 2026-08-22 | 4575 |
| 2026-08-24 | 4578 |
| 2026-08-26 | 4579 |
| 2026-08-27 | 4579 |
| 2026-08-28 | 4580 |
| 2026-08-30 | 4580 |
| 2026-08-31 | 4585 |
| 2026-09-01 | 4587 |
| 2026-09-02 | 4588 |
| 2026-09-03 | 4587 |
| 2026-09-04 | 4588 |
| 2026-09-05 | 4590 |
| 2026-09-07 | 4593 |
| 2026-09-08 | 4597 |
| 2026-09-09 | 4598 |
| 2026-09-10 | 4599 |
| 2026-09-11 | 4601 |
| 2026-09-12 | 4603 |
| 2026-09-13 | 4605 |
| 2026-09-14 | 4608 |
| 2026-09-15 | 4609 |
| 2026-09-16 | 4610 |
| 2026-09-17 | 4610 |
| 2026-09-19 | 4609 |
| 2026-09-21 | 4611 |
| 2026-09-22 | 4613 |
| 2026-09-23 | 4614 |
| 2026-09-24 | 4613 |
| 2026-09-25 | 4614 |
| 2026-09-26 | 4615 |
| 2026-09-27 | 4616 |
| 2026-09-28 | 4615 |
| 2026-09-29 | 4615 |
| 2026-09-30 | 4614 |
| 2026-10-02 | 4616 |
| 2026-10-04 | 4616 |
| 2026-10-05 | 4615 |
| 2026-10-07 | 4616 |
| 2026-10-08 | 4617 |
| 2026-10-09 | 4620 |
| 2026-10-10 | 4619 |
| 2026-10-11 | 4620 |
| 2026-10-12 | 4622 |
| 2026-10-13 | 4623 |
| 2026-10-14 | 4624 |
| 2026-10-15 | 4626 |
| 2026-10-16 | 4627 |
| 2026-10-17 | 4628 |
| 2026-10-18 | 4629 |
| 2026-10-19 | 4630 |
| 2026-10-20 | 4631 |
| 2026-10-22 | 4635 |
| 2026-10-24 | 4635 |
| 2026-10-27 | 4636 |
| 2026-10-28 | 4635 |
| 2026-10-29 | 4636 |
| 2026-10-30 | 4636 |
| 2026-10-31 | 4638 |
| 2026-11-01 | 4639 |
| 2026-11-02 | 4641 |
| 2026-11-03 | 4643 |
| 2026-11-04 | 4644 |
Today
+1 stars today
This week
+9 stars this week
This month
+29 stars this month
Momentum
0.0
growth rate 0.19%/day
<div align="center"> # BEVFormer: a Cutting-edge Baseline for Camera-based Detection </div> https://user-images.githubusercontent.com/27915819/161392594-fc0082f7-5c37-4919-830a-2dd423c1d025.mp4 > **BEVFormer: Learning Bird's-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformers**, ECCV 2022 > - [Paper in arXiv](http://arxiv.org/abs/2203.17270) | [Paper in Chinese](https://drive.google.com/file/d/1dKnD6gUHhBXZ8gT733cIU_A7dHEEzNTP/view?usp=sharing) | [OpenDriveLab](https://opendrivelab.com/) > - [Slides in English](https://docs.google.com/presentation/d/1fTjuSKpj_-KRjUACr8o5TbKetAXlWrmpaorVvfCTZUA/edit?usp=sharing) | [Occupancy and BEV Perception Talk Slides](https://docs.google.com/presentation/d/1U7wVi2_zJxM-EMqLVqC4zJ12ItUgS7ZcsXp7zQ1fkvc/edit?usp=sharing) > - [Blog in Chinese](https://www.zhihu.com/question/521842610/answer/2431585901) | [Video Talk](https://www.bilibili.com/video/BV12t4y1t7Lq?share_source=copy_web) and [Slides](https://docs.google.com/presentation/d/1NNeikhDPkgT14G1D_Ih7K3wbSN0DkvhO9wlAMx3CIcM/edit?usp=sharing) (in Chinese) > - [BEV Perception Survey](https://arxiv.org/abs/2209.05324) (Accepted by PAMI) | [Github repo](https://github.com/OpenDriveLab/BEVPerception-Survey-Recipe) # News - [2022/6/16]: We added two BEVformer configurations, which require less GPU memory than the base version. Please pull this repo to obtain the latest codes. - [2022/6/13]: We release an initial version of BEVFormer. It achieves a baseline result of **51.7%** NDS on nuScenes. - [2022/5/23]: 🚀🚀Built on top of BEVFormer, **BEVFormer++**, gathering up all best practices in recent SOTAs and our unique modification, ranks **1st** on [Waymo Open Datast 3D Camera-Only Detection Challenge](https://waymo.com/open/challenges/2022/3d-camera-only-detection/). We will present BEVFormer++ on CVPR 2022 Autonomous Driving [Workshop](https://cvpr2022.wad.vision/). - [2022/3/10]: 🚀BEVFormer achieve the SOTA on [nuScenes Detection Task](https://nuscenes.org/object-detection?externalData=all&mapData=all&modalities=Camera) with **56.9% NDS** (camera-only)! </br> # Abstract In this work, the authors present a new framework termed BEVFormer, which learns unified BEV representations with spatiotemporal transformers to support multiple autonomous driving perception tasks. In a nutshell, BEVFormer exploits both spatial and temporal information by interacting with spatial and temporal space through predefined grid-shaped BEV queries. To aggregate spatial information, the authors design a spatial cross-attention that each BEV query extracts the spatial features from the regions of interest across camera views. For temporal information, the authors propose a temporal self-attention to recurrently fuse the history BEV information. The proposed approach achieves the new state-of-the-art **56.9\%** in terms of NDS metric on the nuScenes test set, which is **9.0** points higher than previous best arts and on par with the performance of LiDAR-based baselines. # Methods  # Getting Started - [Installation](docs/install.md) - [Prepare Dataset](docs/prepare_dataset.md) - [Run and Eval](docs/getting_started.md) # Model Zoo | Backbone | Method | Lr Schd | NDS| mAP|memroy | Config | Download | | :---: | :---: | :---: | :---: | :---:|:---:| :---: | :---: | | R50 | BEVFormer-tiny_fp16 | 24ep | 35.9|25.7 | - |[config](projects/configs/bevformer_fp16/bevformer_tiny_fp16.py) |[model](https://github.com/zhiqi-li/storage/releases/download/v1.0/bevformer_tiny_fp16_epoch_24.pth)/[log](https://github.com/zhiqi-li/storage/releases/download/v1.0/bevformer_tiny_fp16_epoch_24.log) | | R50 | BEVFormer-tiny | 24ep | 35.4|25.2 | 6500M |[config](projects/configs/bevformer/bevformer_tiny.py) |[model](https://github.com/zhiqi-li/storage/releases/download/v1.0/bevformer_tiny_epoch_24.pth)/[log](https://github.com/zhiqi-li/storage/releases/download/v1.0/bevformer_tiny_epoch_24.log) | | [R101-DCN](h
Excerpt of 8,794 characters
Read on GitHub
Zhiqi Li
12
Chonghao Sima · China
10
Hongyang Li · OpenDriveLab at The University of Hong Kong · Hong Kong
3
2
1
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:39cae233bb2cc51a, topic:computer-vision, topic:object-detection, desc:object detection
matched fp:39cae233bb2cc51a, topic:autonomous-driving, desc:autonomous driving, readme:autonomous driving