RecStitchNet:Learning to stitch images with rectangular boundaries
作者机构：Key Lab of Film and TV Media Technology of Zhejiang ProvinceCollege of Media EngineeringCommunication University of ZhejiangHangzhou 310018China School of Computer Science and InformaticsCardiff UniversityCardiff CF244AGUK Institute of Information ScienceBeijing Jiaotong UniversityBeijing 100091China School of Engineering and Computer ScienceVictoria University of WellingtonWellington 6012New Zealand STEMUniversity of South AustraliaAdelaide 5095Australia
出 版 物：《Computational Visual Media》 (计算可视媒体（英文版）)
年 卷 期：2024年第10卷第4期
页 面：687-703页
核心收录：
学科分类：08[工学] 0812[工学-计算机科学与技术（可授工学、理学学位）]
基 金：supported by the Zhejiang Province Basic Public Welfare Research Program(No.LGG22F020009) Key Lab of Film and TV Media Technology of Zhejiang Province(No.2020E10015) Marsden Fund Council managed by the Royal Society of New Zealand(No.MFP-20-VUW-180)
主 题：image stitching boundaries convolutional neural network
摘 要：Irregular boundaries in image stitching naturally occur due to freely moving *** deal with this problem,existing methods focus on optimizing mesh warping to make boundaries regular using the traditional explicit ***,previous methods always depend on hand-crafted features(e.g.,keypoints and line segments).Thus,failures often happen in overlapping regions without distinctive *** this paper,we address this problem by proposing RecStitchNet,a reasonable and effective network for image stitching with rectangular *** that both stitching and imposing rectangularity are non-trivial tasks in the learning-based framework,we propose a three-step progressive learning based strategy,which not only simplifies this task,but gradually achieves a good balance between stitching and imposing *** the first step,we perform initial stitching by a pre-trained state-of-the-art image stitching model,to produce initially warped stitching results without considering the boundary ***,we use a regression network with a comprehensive objective regarding mesh,perception,and shape to further encourage the stitched meshes to have rectangular boundaries with high content ***,we propose an unsupervised instance-wise optimization strategy to refine the stitched meshes iteratively,which can effectively improve the stitching results in terms of feature alignment,as well as boundary and structure *** to the lack of stitching datasets and the difficulty of label generation,we propose to generate a stitching dataset with rectangular stitched images as pseudo-ground-truth labels,and the performance upper bound induced from the it can be broken by our unsupervised *** and quantitative results and evaluations demonstrate the advantages of our method over the state-of-the-art.
维普期刊数据库