Can I read Modeling Long-range Dependencies and Epipolar Geometry for Multi-view Stereo on EtoBox?

Modeling Long-range Dependencies and Epipolar Geometry for Multi-view Stereo by Jie Zhu; Bo Peng; Wanqing Li; Haifeng Shen; Qingming Huang; Jianjun Lei is a Computer Science article available to read on EtoBox.

What is Modeling Long-range Dependencies and Epipolar Geometry for Multi-view Stereo about?

This article proposes a network, referred to as Multi-View Stereo TRansformer (MVSTR) for depth estimation from multi-view images. By modeling long-range dependencies and epipolar geometry, the proposed MVSTR is capable of extracting dense features with global context and 3D consistency, which are crucial for reliable matching in multi-view stereo (MVS). Specifically, to tackle the problem of the limited receptive field of existing CNN-based MVS methods, a global-context Transformer module is designed to establish intra-view long-range dependencies so that global contextual features of each view are obtained. In addition, to further enable features of each view to be 3D consistent, a 3D-consistency Transformer module with an epipolar feature sampler is built, where epipolar geometry is modeled to effectively facilitate cross-view interaction. Experimental results show that the proposed MVSTR achieves the best overall performance on the DTU dataset and demonstrates strong generalization on the Tanks & Temples benchmark dataset.

Who reads Modeling Long-range Dependencies and Epipolar Geometry for Multi-view Stereo?

Author

Jie Zhu; Bo Peng; Wanqing Li; Haifeng Shen; Qingming Huang; Jianjun Lei

Publisher

ACM

Published

2023

Language

EN

Field

Computer Science (Physical Sciences)