Learning edge-preserved image stitching from multi-scale deep homography

Article Computer Science, Software Engineering

Content-Preserving Image Stitching With Piecewise Rectangular Boundary Constraints

Yun Zhang et al.

Summary: This article proposes an approach to content-preserving image stitching with regular boundary constraints, aiming to generate a panoramic image with piecewise rectangular boundaries. By formulating image stitching with regular boundaries in a unified optimization framework, the method efficiently produces visually pleasing panoramas with regular boundaries and unnoticeable distortions.

IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS (2021)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Image stitching via deep homography estimation

Qiang Zhao et al.

Summary: In this paper, a new deep neural network for image stitching with small parallax images is proposed, with key components designed to improve performance and a new loss function introduced. Experimental results demonstrate that the method outperforms existing methods in terms of quantitative evaluation, visual stitching result, and robustness.

NEUROCOMPUTING (2021)

添加到收藏夹

Proceedings Paper Computer Science, Artificial Intelligence

Motion Basis Learning for Unsupervised Deep Homography Estimation with Subspace Projection

Nianjin Ye et al.

Summary: This paper introduces a new framework for unsupervised deep homography estimation, which includes homography flow representation, Low Rank Representation (LRR) block, and Feature Identity Loss (FIL). These components help achieve more effective and stable unsupervised optimization.

2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021) (2021)

添加到收藏夹

Article Engineering, Electrical & Electronic