Single-Eye View: Monocular Real-time Perception Package for Autonomous Driving

TL;DR

This paper presents LRHPerception, a real-time monocular perception system for autonomous driving that combines efficiency and detailed environmental understanding, achieving high speed and accuracy in processing camera data.

Contribution

The paper introduces LRHPerception, a novel perception package that integrates object detection, road segmentation, and depth estimation into a unified, efficient framework for autonomous driving.

Findings

Processes monocular images into a five-channel tensor including RGB, segmentation, and depth.

Achieves real-time processing at 29 FPS on a single GPU.

Provides a 555% speedup over previous mapping-based approaches.

Abstract

Amidst the rapid advancement of camera-based autonomous driving technology, effectiveness is often prioritized with limited attention to computational efficiency. To address this issue, this paper introduces LRHPerception, a real-time monocular perception package for autonomous driving that uses single-view camera video to interpret the surrounding environment. The proposed system combines the computational efficiency of end-to-end learning with the rich representational detail of local mapping methodologies. With significant improvements in object tracking and prediction, road segmentation, and depth estimation integrated into a unified framework, LRHPerception processes monocular image data into a five-channel tensor consisting of RGB, road segmentation, and pixel-level depth estimation, augmented with object detection and trajectory prediction. Experimental results demonstrate strong performance, achieving real-time processing at 29 FPS on a single GPU, representing a 555% speedup over the fastest mapping-based approach.