Multi-task approaches to joint depth and segmentation prediction are well-studied for monocular images. Yet, predictions from a single-view are inherently limited, while multiple views are available in many robotics applications. On the other end of the spectrum, video-based and full 3D methods require numerous frames to perform reconstruction and segmentation. With this work we propose a Multi-View Stereo (MVS) technique for depth prediction that benefits from rich semantic features of the Segment Anything Model (SAM). This enhanced depth prediction, in turn, serves as a prompt to our Transformer-based semantic segmentation decoder. We report the mutual benefit that both tasks enjoy in our quantitative and qualitative studies on the ScanNet dataset. Our approach consistently outperforms single-task MVS and segmentation models, along with multi-task monocular methods.
@article{arxiv.2311.00134,
title = {Joint Depth Prediction and Semantic Segmentation with Multi-View SAM},
author = {Mykhailo Shvets and Dongxu Zhao and Marc Niethammer and Roni Sengupta and Alexander C. Berg},
journal= {arXiv preprint arXiv:2311.00134},
year = {2023}
}
Comments
To appear in the 2024 IEEE/CVF Winter Conference on Applications of Computer Vision