Generalizable Novel-View Synthesis using a Stereo Camera
This work addresses the challenge of synthesizing novel views from stereo images for applications in computer vision, representing an incremental improvement by combining existing techniques.
The paper tackles the problem of generalizable novel-view synthesis from stereo-camera images by integrating stereo matching into a NeRF-based framework, resulting in StereoNeRF, which surpasses previous approaches in performance.
In this paper, we propose the first generalizable view synthesis approach that specifically targets multi-view stereo-camera images. Since recent stereo matching has demonstrated accurate geometry prediction, we introduce stereo matching into novel-view synthesis for high-quality geometry reconstruction. To this end, this paper proposes a novel framework, dubbed StereoNeRF, which integrates stereo matching into a NeRF-based generalizable view synthesis approach. StereoNeRF is equipped with three key components to effectively exploit stereo matching in novel-view synthesis: a stereo feature extractor, a depth-guided plane-sweeping, and a stereo depth loss. Moreover, we propose the StereoNVS dataset, the first multi-view dataset of stereo-camera images, encompassing a wide variety of both real and synthetic scenes. Our experimental results demonstrate that StereoNeRF surpasses previous approaches in generalizable view synthesis.