02-BEV2OCC.md

BEV

环视BEV感知：下一代自动驾驶感知算法新范式，BEVFormer、BEVFormer++、BEVFusion等系列算法 迈向BEV的大一统 | 基于纯视觉策略的BEV感知

LSS

• https://nv-tlabs.github.io/lift-splat-shoot/
• https://arxiv.org/pdf/2008.05711.pdf

BEVFormer

• https://github.com/fundamentalvision/BEVFormer
• 论文一作李志琦解读BEVFormer： 基于Transformer的自动驾驶BEV纯视觉感知

BEVDet

• https://github.com/HuangJunJie2017/BEVDet

OCC

OccFormer

• https://github.com/zhangyp15/OccFormer

MonoScene

• https://github.com/astra-vision/MonoScene
• 17-MonoScene: 单目3D语义场景补全

Occ3D

• https://github.com/Tsinghua-MARS-Lab/Occ3D

FB-OCC

• :video: CVPR23 | 3D Occupancy 预测冠军方案：FB-OCC
• https://github.com/NVlabs/FB-BEV

SurroundOcc

• https://github.com/weiyithu/SurroundOcc

补充材料：

• :video: Tesla AI Day 2022 - Occupancy

• 纯干货 | Occupancy Networks的个人学习总结！占用网络未来走向如何？

• 3D占据网络的个人学习总结！占用网络未来走向如何？

• 3D视觉Occupancy方法VoxFormer代码详细解读

• Occupancy Network 系列论文汇总

• nuScenes SOTA！SurroundOcc：面向自动驾驶的纯视觉3D占据预测网络（清华&天大）

• CVPR 2023 I VoxFromer: 单目3D语义场景补全新SOTA!

• 场景即占用 | Scene as Occupancy：占用网络新方案！

• 基于视觉的3D Occupancy论文阅读：MonoScene, OccDepth, VoxFormer, TPVFormer

• 【论文阅读】Symphonies(Scene-from-Insts)

• OpenScene 3D Occupancy Prediction Benchmark in Autonomous Driving

• AD23Challenge: 3D Occupancy Prediction

• https://github.com/CVPR2023-3D-Occupancy-Prediction/CVPR2023-3D-Occupancy-Prediction

• occupancy-for-nuscenes Occupancy Dataset for nuScenes