Domain-invariant Stereo Matching Networks

Abstract

State-of-the-art stereo matching networks have difficulties in generalizing to new unseen environments due to significant domain differences, such as color, illumination, contrast, and texture. In this paper, we aim at designing a domain-invariant stereo matching network (DSMNet) that generalizes well to unseen scenes. To achieve this goal, we propose i) a novel "domain normalization" approach that regularizes the distribution of learned representations to allow them to be invariant to domain differences, and ii) a trainable non-local graph-based filter for extracting robust structural and geometric representations that can further enhance domain-invariant generalizations. When trained on synthetic data and generalized to real test sets, our model performs significantly better than all state-of-the-art models. It even outperforms some deep learning models (e.g. MC-CNN) fine-tuned with test-domain data.

References in corpus (7)

• Instance Normalization: The Missing Ingredient for Fast Stylization
• Computing the Stereo Matching Cost with a Convolutional Neural Network
• Understanding the Effective Receptive Field in Deep Convolutional Neural Networks
• CARLA: An Open Urban Driving Simulator
• Self-Supervised Learning for Stereo Matching with Self-Improving Ability
• Anytime Stereo Image Depth Estimation on Mobile Devices
• SegStereo: Exploiting Semantic Information for Disparity Estimation

Cited by in corpus (4)

• AdaStereo: A Simple and Efficient Approach for Adaptive Stereo Matching
• CFNet: Cascade and Fused Cost Volume for Robust Stereo Matching
• Bilateral Grid Learning for Stereo Matching Networks
• Semi-synthesis: A fast way to produce effective datasets for stereo matching