Enregistré dans:
Détails bibliographiques
Auteurs principaux: Zhou, Tianhao, Li, Haipeng, Wang, Ziyi, Luo, Ao, Zhang, Chen-Lin, Li, Jiajun, Zeng, Bing, Liu, Shuaicheng
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:https://arxiv.org/abs/2403.19164
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866911818339844096
author Zhou, Tianhao
Li, Haipeng
Wang, Ziyi
Luo, Ao
Zhang, Chen-Lin
Li, Jiajun
Zeng, Bing
Liu, Shuaicheng
author_facet Zhou, Tianhao
Li, Haipeng
Wang, Ziyi
Luo, Ao
Zhang, Chen-Lin
Li, Jiajun
Zeng, Bing
Liu, Shuaicheng
contents Image stitching from different captures often results in non-rectangular boundaries, which is often considered unappealing. To solve non-rectangular boundaries, current solutions involve cropping, which discards image content, inpainting, which can introduce unrelated content, or warping, which can distort non-linear features and introduce artifacts. To overcome these issues, we introduce a novel diffusion-based learning framework, \textbf{RecDiffusion}, for image stitching rectangling. This framework combines Motion Diffusion Models (MDM) to generate motion fields, effectively transitioning from the stitched image's irregular borders to a geometrically corrected intermediary. Followed by Content Diffusion Models (CDM) for image detail refinement. Notably, our sampling process utilizes a weighted map to identify regions needing correction during each iteration of CDM. Our RecDiffusion ensures geometric accuracy and overall visual appeal, surpassing all previous methods in both quantitative and qualitative measures when evaluated on public benchmarks. Code is released at https://github.com/lhaippp/RecDiffusion.
format Preprint
id arxiv_https___arxiv_org_abs_2403_19164
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle RecDiffusion: Rectangling for Image Stitching with Diffusion Models
Zhou, Tianhao
Li, Haipeng
Wang, Ziyi
Luo, Ao
Zhang, Chen-Lin
Li, Jiajun
Zeng, Bing
Liu, Shuaicheng
Computer Vision and Pattern Recognition
Image stitching from different captures often results in non-rectangular boundaries, which is often considered unappealing. To solve non-rectangular boundaries, current solutions involve cropping, which discards image content, inpainting, which can introduce unrelated content, or warping, which can distort non-linear features and introduce artifacts. To overcome these issues, we introduce a novel diffusion-based learning framework, \textbf{RecDiffusion}, for image stitching rectangling. This framework combines Motion Diffusion Models (MDM) to generate motion fields, effectively transitioning from the stitched image's irregular borders to a geometrically corrected intermediary. Followed by Content Diffusion Models (CDM) for image detail refinement. Notably, our sampling process utilizes a weighted map to identify regions needing correction during each iteration of CDM. Our RecDiffusion ensures geometric accuracy and overall visual appeal, surpassing all previous methods in both quantitative and qualitative measures when evaluated on public benchmarks. Code is released at https://github.com/lhaippp/RecDiffusion.
title RecDiffusion: Rectangling for Image Stitching with Diffusion Models
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2403.19164