Multi-View Stereo with Transformer
This work addresses the problem of reliable 3D reconstruction from multiple images for computer vision applications, presenting a novel method that improves upon existing approaches.
The paper tackles the limited receptive field and lack of 3D consistency in CNN-based Multi-View Stereo methods by proposing MVSTR, a Transformer-based network that achieves state-of-the-art performance on the DTU dataset and strong generalization on Tanks & Temples.
This paper proposes a network, referred to as MVSTR, for Multi-View Stereo (MVS). It is built upon Transformer and is capable of extracting dense features with global context and 3D consistency, which are crucial to achieving reliable matching for MVS. Specifically, to tackle the problem of the limited receptive field of existing CNN-based MVS methods, a global-context Transformer module is first proposed to explore intra-view global context. In addition, to further enable dense features to be 3D-consistent, a 3D-geometry Transformer module is built with a well-designed cross-view attention mechanism to facilitate inter-view information interaction. Experimental results show that the proposed MVSTR achieves the best overall performance on the DTU dataset and strong generalization on the Tanks & Temples benchmark dataset.