Geometric Reciprocity: Unlocking Self-Supervision for Stereoscopic Video Generation

Monocular-to-stereo conversion synthesizes stereoscopic content from 2D videos for immersive 3D experiences. In modern Depth-Image-Based Rendering (DIBR) approaches, stereo inpainting of disocclusions is the critical bottleneck. Training-based methods achieve superior quality but rely on scarce stereo pairs or synthetic data with domain gaps. We address this through the first self-supervised framework learning from monocular videos via cycle consistency. Our key contribution is the Geometric Reciprocity Theorem (GRT): under the nearest-neighbor DIBR formulation, the disocclusion mask when synthesizing a target view equals the mask of pixels lost when warping back from target to source, enabling analytical computation of test-time disocclusion masks directly from monocular images. This yields train-test consistency for the stated warping formulation, supporting self-supervised learning from unlimited monocular videos and substantial improvements over training-free and supervised state-of-the-art methods. Project page: https://visual-ai.github.io/grt/

This paper studies how to convert ordinary 2D videos into stereoscopic 3D videos. A key challenge is that when a new eye view is created, some regions become newly visible and must be filled in naturally. Existing learning-based methods usually need stereo video pairs or synthetic data, which are hard to obtain or may not match real videos well. We introduce Geometric Reciprocity, a way to automatically create training masks from ordinary single-view images and videos using geometry. This makes it possible to train stereo inpainting models from large collections of regular videos, improving the quality and consistency of generated 3D content.