Research map / Vision-language and multimodal / Cross-modal learning
HOMO-Feature: Cross-Arbitrary-Modal Image Matching with Homomorphism of Organized Major Orientation
Chenzhong Gao, Wei Li, Desheng Weng
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace non-data-driven HOMO with deep matchers tuned on cross-modal data (self-tuned LoFTR) | LoFTR · GCZ feature invariance (NCM) | NCM | Improved |
| tune LoFTR on rotated datasets to gain rotation invariance | LoFTR · rotated cross-modal matching | matching performance | No clear effect |
| remove multi-scale strategy (HOMO-single) (ablation) | HOMO · GCZ feature invariance (NCM) | NCM | Hurt |
| add OCD (orientation consistency) within GPolar descriptor (ablation) | HOMO · large angular differences | NCM | Improved |
| add deep form of GPolar (ablation) | HOMO with multi-scale strategy · average NCM | NCM | No clear effect |
| add deep form of GPolar to single-scale HOMO (ablation) | HOMO-single · feature stability | NCM | Improved |
| use MOM and GPolar instead of gradient features or HOG/GLOH/LogPolar (ablation) | HOMO · cross-modal matching | NCM | Improved |
| use HOMO instead of deep networks on multi-view 3D scenes | HOMO · METU-VisTIR multi-view | global matches | Hurt |
Closest papers by meaning
- HIMO: Cross-Arbitrary-Modality Image Invariant Feature Transform With Hierarchical Intrinsic Major Orientation (TPAMI 2026 · similarity 0.91)
- SCPNet: Unsupervised Cross-modal Homography Estimation via Intra-modal Self-supervised Learning (ECCV 2024 · similarity 0.88)
- SSHNet: Unsupervised Cross-modal Homography Estimation via Problem Reformulation and Split Optimization (CVPR 2025 · similarity 0.86)
- Self-Supervised Spatial Correspondence Across Modalities (CVPR 2025 · similarity 0.84)
- Adapting Dense Matching for Homography Estimation with Grid-based Acceleration (CVPR 2025 · similarity 0.84)
- Unsupervised Homography Estimation on Multimodal Image Pair via Alternating Optimization (NeurIPS 2024 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
