Towards Open-World Generation of Stereo Images and Unsupervised Matching
This addresses the problem of acquiring precise stereo data for researchers and engineers in fields requiring 3D perception, though it appears incremental as it builds on existing diffusion methods with specific enhancements.
The paper tackles the challenge of generating high-quality stereo images that are both visually realistic and geometrically accurate for applications like XR and autonomous driving, introducing GenStereo, a diffusion-based method that achieves state-of-the-art performance in stereo image generation and unsupervised matching.
Stereo images are fundamental to numerous applications, including extended reality (XR) devices, autonomous driving, and robotics. Unfortunately, acquiring high-quality stereo images remains challenging due to the precise calibration requirements of dual-camera setups and the complexity of obtaining accurate, dense disparity maps. Existing stereo image generation methods typically focus on either visual quality for viewing or geometric accuracy for matching, but not both. We introduce GenStereo, a diffusion-based approach, to bridge this gap. The method includes two primary innovations (1) conditioning the diffusion process on a disparity-aware coordinate embedding and a warped input image, allowing for more precise stereo alignment than previous methods, and (2) an adaptive fusion mechanism that intelligently combines the diffusion-generated image with a warped image, improving both realism and disparity consistency. Through extensive training on 11 diverse stereo datasets, GenStereo demonstrates strong generalization ability. GenStereo achieves state-of-the-art performance in both stereo image generation and unsupervised stereo matching tasks. Project page is available at https://qjizhi.github.io/genstereo.