AdaStereo: An Efficient Domain-Adaptive Stereo Matching Approach

TL;DR

AdaStereo introduces a comprehensive domain-adaptive stereo matching method that aligns multi-level representations through input, feature, and output space techniques, achieving state-of-the-art results without extra inference costs.

Contribution

The paper presents a novel, complete domain adaptation pipeline for stereo matching, including a non-adversarial color transfer, parameter-free feature normalization, and self-supervised occlusion-aware reconstruction.

Findings

Achieves state-of-the-art cross-domain performance on multiple benchmarks.

Outperforms some models finetuned with target ground-truths.

Demonstrates robustness and easy integration into real-world scenarios.

Abstract

Recently, records on stereo matching benchmarks are constantly broken by end-to-end disparity networks. However, the domain adaptation ability of these deep models is quite limited. Addressing such problem, we present a novel domain-adaptive approach called AdaStereo that aims to align multi-level representations for deep stereo matching networks. Compared to previous methods, our AdaStereo realizes a more standard, complete and effective domain adaptation pipeline. Firstly, we propose a non-adversarial progressive color transfer algorithm for input image-level alignment. Secondly, we design an efficient parameter-free cost normalization layer for internal feature-level alignment. Lastly, a highly related auxiliary task, self-supervised occlusion-aware reconstruction is presented to narrow the gaps in output space. We perform intensive ablation studies and break-down comparisons to validate the effectiveness of each proposed module. With no extra inference overhead and only a slight increase in training complexity, our AdaStereo models achieve state-of-the-art cross-domain performance on multiple benchmarks, including KITTI, Middlebury, ETH3D and DrivingStereo, even outperforming some state-of-the-art disparity networks finetuned with target-domain ground-truths. Moreover, based on two additional evaluation metrics, the superiority of our domain-adaptive stereo matching pipeline is further uncovered from more perspectives. Finally, we demonstrate that our method is robust to various domain adaptation settings, and can be easily integrated into quick adaptation application scenarios and real-world deployments.