3D semantic occupancy prediction is an essential part of autonomous driving, focusing on capturing the geometric details of scenes. Off-road environments are rich in geometric information, therefore it is suitable for 3D semantic occupancy prediction tasks to reconstruct such scenes. However, most of researches concentrate on on-road environments, and few methods are designed for off-road 3D semantic occupancy prediction due to the lack of relevant datasets and benchmarks. In response to this gap, we introduce WildOcc, to our knowledge, the first benchmark to provide dense occupancy annotations for off-road 3D semantic occupancy prediction tasks. A ground truth generation pipeline is proposed in this paper, which employs a coarse-to-fine reconstruction to achieve a more realistic result. Moreover, we introduce a multi-modal 3D semantic occupancy prediction framework, which fuses spatio-temporal information from multi-frame images and point clouds at voxel level. In addition, a cross-modality distillation function is introduced, which transfers geometric knowledge from point clouds to image features.
@article{arxiv.2410.15792,
title = {WildOcc: A Benchmark for Off-Road 3D Semantic Occupancy Prediction},
author = {Heng Zhai and Jilin Mei and Chen Min and Liang Chen and Fangzhou Zhao and Yu Hu},
journal= {arXiv preprint arXiv:2410.15792},
year = {2024}
}