LiSD: An Efficient Multi-Task Learning Framework for LiDAR Segmentation and Detection

TL;DR

LiSD is a novel multi-task learning framework that efficiently combines lidar segmentation and detection tasks, achieving state-of-the-art results by leveraging hierarchical feature collaboration and cross-task information sharing.

Contribution

This work introduces LiSD, the first efficient multi-task framework for lidar segmentation and detection that integrates hierarchical and holistic modules for improved performance.

Findings

Achieves 83.3% mIoU on nuScenes segmentation benchmark.

Outperforms existing methods on nuScenes and Waymo datasets.

Effectively balances sparsity and density in feature integration.

Abstract

With the rapid proliferation of autonomous driving, there has been a heightened focus on the research of lidar-based 3D semantic segmentation and object detection methodologies, aiming to ensure the safety of traffic participants. In recent decades, learning-based approaches have emerged, demonstrating remarkable performance gains in comparison to conventional algorithms. However, the segmentation and detection tasks have traditionally been examined in isolation to achieve the best precision. To this end, we propose an efficient multi-task learning framework named LiSD which can address both segmentation and detection tasks, aiming to optimize the overall performance. Our proposed LiSD is a voxel-based encoder-decoder framework that contains a hierarchical feature collaboration module and a holistic information aggregation module. Different integration methods are adopted to keep sparsity in segmentation while densifying features for query initialization in detection. Besides, cross-task information is utilized in an instance-aware refinement module to obtain more accurate predictions. Experimental results on the nuScenes dataset and Waymo Open Dataset demonstrate the effectiveness of our proposed model. It is worth noting that LiSD achieves the state-of-the-art performance of 83.3% mIoU on the nuScenes segmentation benchmark for lidar-only methods.