Renderworld: World Model with Self-Supervised 3D Label
Ziyang Yan, Wenzhen Dong, Yihua Shao, Yuhang Lu, Haiyang Liu, Jingwen Liu, Haozhe Wang, Zhe Wang
Abstract
End-to-end autonomous driving with vision-only is not only more cost-effective compared to LiDAR-vision fusion but also more reliable than traditional methods. To achieve a economical and robust purely visual autonomous driving system, we propose RenderWorld, a vision-only end-to-end autonomous driving framework, which generates 3D occupancy labels using a self-supervised gaussian-based Img2Occ Module, then encodes the labels by AM-VAE, and uses world model for forecasting and planning. RenderWorld employs Gaussian Splatting to represent 3D scenes and render 2D images greatly improves segmentation accuracy and reduces GPU memory consumption compared with NeRF-based methods. By applying AM-VAE to encode air and non-air separately, RenderWorld achieves more fine-grained scene element representation, leading to state-of-the-art performance in both 4D occupancy forecasting and motion planning from autoregressive world model.
BibTeX
@inproceedings{icra2025_renderworldworld,
title = {Renderworld: World Model with Self-Supervised 3D Label},
author = {Ziyang Yan and Wenzhen Dong and Yihua Shao and Yuhang Lu and Haiyang Liu and Jingwen Liu and Haozhe Wang and Zhe Wang and Yan Wang and Fabio Remondino and Yuexin Ma},
booktitle = {ICRA 2025},
year = {2025}
}