A Point-Based Approach to Efficient LiDAR Multi-Task Perception

TL;DR

PAttFormer is a novel point-based multi-task architecture for LiDAR perception that is smaller, faster, and maintains competitive accuracy for semantic segmentation and object detection.

Contribution

It introduces a transformer-based, point-only architecture that eliminates the need for multiple task-specific encoders, improving efficiency in LiDAR multi-task perception.

Findings

3x smaller network size compared to previous methods

1.4x faster inference speed

Improved segmentation and detection accuracy on nuScenes

Abstract

Multi-task networks can potentially improve performance and computational efficiency compared to single-task networks, facilitating online deployment. However, current multi-task architectures in point cloud perception combine multiple task-specific point cloud representations, each requiring a separate feature encoder and making the network structures bulky and slow. We propose PAttFormer, an efficient multi-task architecture for joint semantic segmentation and object detection in point clouds that only relies on a point-based representation. The network builds on transformer-based feature encoders using neighborhood attention and grid-pooling and a query-based detection decoder using a novel 3D deformable-attention detection head design. Unlike other LiDAR-based multi-task architectures, our proposed PAttFormer does not require separate feature encoders for multiple task-specific point cloud representations, resulting in a network that is 3x smaller and 1.4x faster while achieving competitive performance on the nuScenes and KITTI benchmarks for autonomous driving perception. Our extensive evaluations show substantial gains from multi-task learning, improving LiDAR semantic segmentation by +1.7% in mIou and 3D object detection by +1.7% in mAP on the nuScenes benchmark compared to the single-task models.