Semi-Supervised Disparity Estimation with Deep Feature Reconstruction

2021/06/01 by Julia Guerrero‐Viu, Guerrero-Viu, Julia, Sergio Izquierdo +5
Computer Science · #Advanced Vision and Imaging #Computer Vision and Pattern Recognition (cs.CV) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Human Pose and Action Recognition

paper · pdf · doi:10.48550/arxiv.2106.00318

openalex publication_date 2021/06/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Despite the success of deep learning in disparity estimation, the domain generalization gap remains an issue. We propose a semi-supervised pipeline that successfully adapts DispNet to a real-world domain by joint supervised training on labeled synthetic data and self-supervised training on unlabeled real data. Furthermore, accounting for the limitations of the widely-used photometric loss, we analyze the impact of deep feature reconstruction as a promising supervisory signal for disparity estimation.

Citations

• Vision meets robotics: The KITTI dataset
• End-to-End Learning of Geometry and Context for Deep Stereo Regression

Related

• Revisiting Disparity from Dual-Pixel Images: Physics-Informed Lightweight Depth Estimation
• EDNet: Efficient Disparity Estimation with Cost Volume Combination and Attention-based Spatial Residual
• Learning the Distribution of Errors in Stereo Matching for Joint Disparity and Uncertainty Estimation
• DispBench: Benchmarking Disparity Estimation to Synthetic Corruptions
• Reversing the cycle: self-supervised deep stereo through enhanced\n monocular distillation