Multi-View Inpainting for RGB-D Sequence
This work addresses a domain-specific issue for 3D reconstruction from RGB-D data, presenting an incremental improvement by combining existing techniques like multi-view inpainting and MRF-based methods.
The paper tackles the problem of removing undesired objects from RGB-D sequences to enable static 3D reconstruction, achieving qualified results for object removal and hole inpainting as shown in experiments.
In this work we propose a novel approach to remove undesired objects from RGB-D sequences captured with freely moving cameras, which enables static 3D reconstruction. Our method jointly uses existing information from multiple frames as well as generates new one via inpainting techniques. We use balanced rules to select source frames; local homography based image warping method for alignment and Markov random field (MRF) based approach for combining existing information. For the left holes, we employ exemplar based multi-view inpainting method to deal with the color image and coherently use it as guidance to complete the depth correspondence. Experiments show that our approach is qualified for removing the undesired objects and inpainting the holes.