Joint Depth Prediction and Semantic Segmentation With Multi-View SAM

Mykhailo Shvets, Dongxu Zhao, Marc Niethammer, Roni Sengupta, Alexander C. Berg; Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2024, pp. 1328-1338

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.