OccupancyDETR: Using DETR for Mixed Dense-sparse 3D Occupancy Prediction

TL;DR

OccupancyDETR introduces a novel 3D occupancy perception method combining DETR-like detection with mixed dense-sparse decoding, achieving faster, resource-efficient, and accurate semantic occupancy prediction suitable for real-time robotics applications.

Contribution

It presents a new approach integrating DETR detection with mixed dense-sparse decoding for 3D occupancy, improving efficiency and small object detection performance.

Findings

Achieves 14 mIoU on SemanticKITTI dataset.

Operates at 10 FPS, demonstrating real-time capability.

Reduces resource consumption compared to traditional methods.

Abstract

Visual-based 3D semantic occupancy perception is a key technology for robotics, including autonomous vehicles, offering an enhanced understanding of the environment by 3D. This approach, however, typically requires more computational resources than BEV or 2D methods. We propose a novel 3D semantic occupancy perception method, OccupancyDETR, which utilizes a DETR-like object detection, a mixed dense-sparse 3D occupancy decoder. Our approach distinguishes between foreground and background within a scene. Initially, foreground objects are detected using the DETR-like object detection. Subsequently, queries for both foreground and background objects are fed into the mixed dense-sparse 3D occupancy decoder, performing upsampling in dense and sparse methods, respectively. Finally, a MaskFormer is utilized to infer the semantics of the background voxels. Our approach strikes a balance between efficiency and accuracy, achieving faster inference times, lower resource consumption, and improved performance for small object detection. We demonstrate the effectiveness of our proposed method on the SemanticKITTI dataset, showcasing an mIoU of 14 and a processing speed of 10 FPS, thereby presenting a promising solution for real-time 3D semantic occupancy perception.