RaySt3R: Predicting Novel Depth Maps for Zero-Shot Object Completion
This work addresses limitations in 3D consistency, computational cost, and boundary sharpness for applications in robotics, digital twin reconstruction, and extended reality.
The paper tackles the problem of 3D shape completion from single RGB-D images by recasting it as a novel view synthesis task, achieving state-of-the-art performance with up to 44% improvement in 3D chamfer distance over baselines.
3D shape completion has broad applications in robotics, digital twin reconstruction, and extended reality (XR). Although recent advances in 3D object and scene completion have achieved impressive results, existing methods lack 3D consistency, are computationally expensive, and struggle to capture sharp object boundaries. Our work (RaySt3R) addresses these limitations by recasting 3D shape completion as a novel view synthesis problem. Specifically, given a single RGB-D image and a novel viewpoint (encoded as a collection of query rays), we train a feedforward transformer to predict depth maps, object masks, and per-pixel confidence scores for those query rays. RaySt3R fuses these predictions across multiple query views to reconstruct complete 3D shapes. We evaluate RaySt3R on synthetic and real-world datasets, and observe it achieves state-of-the-art performance, outperforming the baselines on all datasets by up to 44% in 3D chamfer distance. Project page: https://rayst3r.github.io