paper-with-me

홈 › Papers

Cost Volume Pyramid Network with Multi-strategies Range Searching for Multi-view Stereo

2022-07-25 · Shiyu Gao, Zhaoxin Li, Zhaoqi Wang

Multi-view stereo is an important research task in computer vision while still keeping challenging. In recent years, deep learning-based methods have shown superior performance on this task. Cost volume pyramid network-based methods which progressively refine depth map in coarse-to-fine manner, have yielded promising results while consuming less memory. However, these methods fail to take fully consideration of the characteristics of the cost volumes in each stage, leading to adopt similar range search strategies for each cost volume stage. In this work, we present a novel cost volume pyramid based network with different searching strategies for multi-view stereo. By choosing different depth range sampling strategies and applying adaptive unimodal filtering, we are able to obtain more accurate depth estimation in low resolution stages and iteratively upsample depth map to arbitrary resolution. We conducted extensive experiments on both DTU and BlendedMVS datasets, and results show that our method outperforms most state-of-the-art methods.

📄 PDF Abstract BibTeX arXiv:2207.12032

Code (1)

SibylGao/MSCVP-MVSNet 공식 구현 pytorch

Tasks

3D ReconstructionDepth EstimationStereo Matching

Similar Papers 제목 키워드 기반

Cost Volume Pyramid Based Depth Inference for Multi-View Stereo

2019-12-18 · CVPR 2020 6 · Jiayu Yang, Wei Mao, Jose M. Alvarez, Miaomiao Liu

We propose a cost volume-based neural network for depth inference from multi-view images. We demonstrate that building a cost volume pyramid in a coarse-to-fine manner instead of constructing a cost volume at a fixed res…

3D ReconstructionPoint Clouds

PCW-Net: Pyramid Combination and Warping Cost Volume for Stereo Matching

2020-06-23 · Zhelun Shen, Yuchao Dai, Xibin Song, Zhibo Rao 외

Existing deep learning based stereo matching methods either focus on achieving optimal performances on the target dataset while with poor generalization for other datasets or focus on handling the cross-domain generaliza…

Disparity EstimationDomain GeneralizationStereo Matching

Cascade Cost Volume for High-Resolution Multi-View Stereo and Stereo Matching

2019-12-13 · CVPR 2020 6 · Xiaodong Gu, Zhiwen Fan, Zuozhuo Dai, Siyu Zhu 외

The deep multi-view stereo (MVS) and stereo matching approaches generally construct 3D cost volumes to regularize and regress the output depth or disparity. These methods are limited when high-resolution outputs are need…

3D ReconstructionGPUPoint CloudsStereo Matching

PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volume

2017-09-07 · CVPR 2018 6 · Deqing Sun, Xiaodong Yang, Ming-Yu Liu, Jan Kautz

We present a compact but effective CNN model for optical flow, called PWC-Net. PWC-Net has been designed according to simple and well-established principles: pyramidal processing, warping, and the use of a cost volume. C…

Dense Pixel Correspondence EstimationOptical Flow Estimation

Multi-scale Cross-form Pyramid Network for Stereo Matching

2019-04-25 · Zhidong Zhu, Mingyi He, Yuchao Dai, Zhibo Rao 외

Stereo matching plays an indispensable part in autonomous driving, robotics and 3D scene reconstruction. We propose a novel deep learning architecture, which called CFP-Net, a Cross-Form Pyramid stereo matching network f…

3D Feature Matching3D Scene ReconstructionAutonomous DrivingForm+2