Monocular BEV Perception of Road Scenes via Front-to-Top View Projection

TL;DR

This paper introduces a novel monocular camera-based framework for reconstructing bird's-eye view maps of road scenes, improving accuracy and efficiency over existing methods by using a front-to-top view projection with cycle consistency and multi-scale features.

Contribution

The paper presents a new front-to-top view projection module with cycle consistency and multi-scale features for monocular BEV perception, achieving state-of-the-art results and real-time performance.

Findings

State-of-the-art performance in road layout and vehicle occupancy estimation

Outperforms competitors in multi-class semantic estimation

Runs at 25 FPS on a single GPU

Abstract

HD map reconstruction is crucial for autonomous driving. LiDAR-based methods are limited due to expensive sensors and time-consuming computation. Camera-based methods usually need to perform road segmentation and view transformation separately, which often causes distortion and missing content. To push the limits of the technology, we present a novel framework that reconstructs a local map formed by road layout and vehicle occupancy in the bird's-eye view given a front-view monocular image only. We propose a front-to-top view projection (FTVP) module, which takes the constraint of cycle consistency between views into account and makes full use of their correlation to strengthen the view transformation and scene understanding. In addition, we also apply multi-scale FTVP modules to propagate the rich spatial information of low-level features to mitigate spatial deviation of the predicted object location. Experiments on public benchmarks show that our method achieves the state-of-the-art performance in the tasks of road layout estimation, vehicle occupancy estimation, and multi-class semantic estimation. For multi-class semantic estimation, in particular, our model outperforms all competitors by a large margin. Furthermore, our model runs at 25 FPS on a single GPU, which is efficient and applicable for real-time panorama HD map reconstruction.