Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2505.16599 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913854580064256 |
|---|---|
| author | Huang, Yao Cao, Si-Yuan Ding, Yaqing Yin, Hao Xie, Shibin Wang, Shuting Fang, Zhijun Wang, Jiachun Cai, Shen Yan, Junchi Shen, Shuhan |
| author_facet | Huang, Yao Cao, Si-Yuan Ding, Yaqing Yin, Hao Xie, Shibin Wang, Shuting Fang, Zhijun Wang, Jiachun Cai, Shen Yan, Junchi Shen, Shuhan |
| contents | Planar homography, with eight degrees of freedom (DOFs), is fundamental in numerous computer vision tasks. While the positional offsets of four corners are widely adopted (especially in neural network predictions), this parameterization lacks geometric interpretability and typically requires solving a linear system to compute the homography matrix. This paper presents a novel geometric parameterization of homographies, leveraging the similarity-kernel-similarity (SKS) decomposition for projective transformations. Two independent sets of four geometric parameters are decoupled: one for a similarity transformation and the other for the kernel transformation. Additionally, the geometric interpretation linearly relating the four kernel transformation parameters to angular offsets is derived. Our proposed parameterization allows for direct homography estimation through matrix multiplication, eliminating the need for solving a linear system, and achieves performance comparable to the four-corner positional offsets in deep homography estimation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_16599 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Decoupled Geometric Parameterization and its Application in Deep Homography Estimation Huang, Yao Cao, Si-Yuan Ding, Yaqing Yin, Hao Xie, Shibin Wang, Shuting Fang, Zhijun Wang, Jiachun Cai, Shen Yan, Junchi Shen, Shuhan Computer Vision and Pattern Recognition Planar homography, with eight degrees of freedom (DOFs), is fundamental in numerous computer vision tasks. While the positional offsets of four corners are widely adopted (especially in neural network predictions), this parameterization lacks geometric interpretability and typically requires solving a linear system to compute the homography matrix. This paper presents a novel geometric parameterization of homographies, leveraging the similarity-kernel-similarity (SKS) decomposition for projective transformations. Two independent sets of four geometric parameters are decoupled: one for a similarity transformation and the other for the kernel transformation. Additionally, the geometric interpretation linearly relating the four kernel transformation parameters to angular offsets is derived. Our proposed parameterization allows for direct homography estimation through matrix multiplication, eliminating the need for solving a linear system, and achieves performance comparable to the four-corner positional offsets in deep homography estimation. |
| title | Decoupled Geometric Parameterization and its Application in Deep Homography Estimation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2505.16599 |