TiCoSS: Tightening the Coupling between Semantic Segmentation and Stereo Matching within A Joint Learning Framework

TL;DR

TiCoSS is a novel joint learning framework that tightly couples semantic segmentation and stereo matching, significantly improving performance by introducing feature fusion, deep supervision, and a coupling loss.

Contribution

The paper presents three innovative strategies to tightly integrate semantic segmentation and stereo matching within a single framework, advancing the state-of-the-art in autonomous driving perception.

Findings

Over 9% increase in mIoU on KITTI datasets

State-of-the-art performance in joint segmentation and stereo matching

Effective strategies validated through extensive experiments

Abstract

Semantic segmentation and stereo matching, respectively analogous to the ventral and dorsal streams in our human brain, are two key components of autonomous driving perception systems. Addressing these two tasks with separate networks is no longer the mainstream direction in developing computer vision algorithms, particularly with the recent advances in large vision models and embodied artificial intelligence. The trend is shifting towards combining them within a joint learning framework, especially emphasizing feature sharing between the two tasks. The major contributions of this study lie in comprehensively tightening the coupling between semantic segmentation and stereo matching. Specifically, this study introduces three novelties: (1) a tightly coupled, gated feature fusion strategy, (2) a hierarchical deep supervision strategy, and (3) a coupling tightening loss function. The combined use of these technical contributions results in TiCoSS, a state-of-the-art joint learning framework that simultaneously tackles semantic segmentation and stereo matching. Through extensive experiments on the KITTI and vKITTI2 datasets, along with qualitative and quantitative analyses, we validate the effectiveness of our developed strategies and loss function, and demonstrate its superior performance compared to prior arts, with a notable increase in mIoU by over 9%. Our source code will be publicly available at mias.group/TiCoSS upon publication.