MVSCRF: Learning Multi-View Stereo With Conditional Random Fields

Youze Xue, Jiansheng Chen, Weitao Wan, Yiqing Huang, Cheng Yu, Tianpeng Li, Jiayu Bao; Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2019, pp. 4312-4321

Abstract


We present a deep-learning architecture for multi-view stereo with conditional random fields (MVSCRF). Given an arbitrary number of input images, we first use a U-shape neural network to extract deep features incorporating both global and local information, and then build a 3D cost volume for the reference camera. Unlike previous learning based methods, we explicitly constraint the smoothness of depth maps by using conditional random fields (CRFs) after the stage of cost volume regularization. The CRFs module is implemented as recurrent neural networks so that the whole pipeline can be trained end-to-end. Our results show that the proposed pipeline outperforms previous state-of-the-arts on large-scale DTU dataset. We also achieve comparable results with state-of-the-art learning based methods on outdoor Tanks and Temples dataset without fine-tuning, which demonstrates our method's generalization ability.

Related Material


[pdf]
[bibtex]
@InProceedings{Xue_2019_ICCV,
author = {Xue, Youze and Chen, Jiansheng and Wan, Weitao and Huang, Yiqing and Yu, Cheng and Li, Tianpeng and Bao, Jiayu},
title = {MVSCRF: Learning Multi-View Stereo With Conditional Random Fields},
booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},
month = {October},
year = {2019}
}