CuriousCat-7 / SimpleOccupancy

A Simple Framework for 3D Occupancy Estimation in Autonomous Driving

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

SimpleOccupancy

Arxiv Paper

A Simple Framework for 3D Occupancy Estimation in Autonomous Driving

Wanshui Gan, Ningkai Mo, Hongbin Xu, Naoto Yokoya

News

-- Note that the demos are large, please wait a moment to load them.

  • [2023/10/20]: We extend the framework to the 3D reconstruction task based on the SDF at the mesh level with the self-supervised setting. I am open to discussion and collaboration on related topics.
  • [2023/10/07]: Update the paper. The first and preliminary version is realeased. Code may not be cleaned thoroughly, so feel free to open an issue if any question.
  • [2023/4/05]: Update the paper with supplementary material. Code repository is still under construction.
  • [2023/3/18]: Initial release.

Demo

RGB, Depth and Mesh:

Self-supervised learning with SDF (Max depth = 52 m )


Self-supervised learning with Density (Max depth = 52 m)

Sparse occupancy prediction:

Dense occupancy prediction:

Point-level training as the pretrain for 3D semantic occupancy:

Abstract

The task of estimating 3D occupancy from surrounding-view images is an exciting development in the field of autonomous driving, following the success of Bird's Eye View (BEV) perception. This task provides crucial 3D attributes of the driving environment, enhancing the overall understanding and perception of the surrounding space. In this work, we present a simple framework for 3D occupancy estimation, which is a CNN-based framework designed to reveal several key factors for 3D occupancy estimation, such as network design, optimization, and evaluation. In addition, we explore the relationship between 3D occupancy estimation and other related tasks, such as monocular depth estimation and 3D reconstruction, which could advance the study of 3D perception in autonomous driving. For evaluation, we propose a simple sampling strategy to define the metric for occupancy evaluation, which is flexible for current public datasets. Moreover, we establish the benchmark in terms of the depth estimation metric, where we compare our proposed method with monocular depth estimation methods on the DDAD and Nuscenes datasets and achieve competitive performance.

Method

Proposed network:

Occupancy label and metric comparison:

Acknowledgement

Many thanks to these excellent projects:

Related Projects:

Bibtex

If you find this repository/work helpful in your research, welcome to cite the paper and give a ⭐.

@article{gan2023simple,
  title={A Simple Attempt for 3D Occupancy Estimation in Autonomous Driving},
  author={Gan, Wanshui and Mo, Ningkai and Xu, Hongbin and Yokoya, Naoto},
  journal={arXiv preprint arXiv:2303.10076},
  year={2023}
}

About

A Simple Framework for 3D Occupancy Estimation in Autonomous Driving


Languages

Language:Python 99.9%Language:Shell 0.1%