Rethinking Iterative Stereo Matching from Diffusion Bridge Model Perspective

TL;DR

This paper introduces a novel diffusion model-based iterative stereo matching approach that enhances disparity map detail and accuracy, achieving state-of-the-art results with fewer iterations.

Contribution

It proposes integrating diffusion models into stereo matching with a new T-GRU and attention mechanisms, improving detail preservation and performance.

Findings

Achieved first place on Scene Flow dataset

Over 7% performance improvement over competitors

Requires only 8 iterations for top results

Abstract

Recently, iteration-based stereo matching has shown great potential. However, these models optimize the disparity map using RNN variants. The discrete optimization process poses a challenge of information loss, which restricts the level of detail that can be expressed in the generated disparity map. In order to address these issues, we propose a novel training approach that incorporates diffusion models into the iterative optimization process. We designed a Time-based Gated Recurrent Unit (T-GRU) to correlate temporal and disparity outputs. Unlike standard recurrent units, we employ Agent Attention to generate more expressive features. We also designed an attention-based context network to capture a large amount of contextual information. Experiments on several public benchmarks show that we have achieved competitive stereo matching performance. Our model ranks first in the Scene Flow dataset, achieving over a 7% improvement compared to competing methods, and requires only 8 iterations to achieve state-of-the-art results.