Research map / Detection and segmentation / CNN features
Corner Detection Using Second-Order Generalized Gaussian Directional Derivative Representations
Weichuan Zhang, Changming Sun
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 |
|---|---|---|---|
| use SOGGDD proposed detector over ACJ and ANDD | corner detection · 'Geometric' image | total missed plus false corners: 28 -> 18 | Improved |
| use SOGGDD proposed detector over ANDD | corner detection · 'Lab' image | total missed plus false corners: 156 -> 141 | Improved |
| use SOGGDD proposed detector over Harris | corner detection · 'Geometric' image | localization error (pixels): 1.269 -> 1.177 | Improved |
| use SOGGDD proposed detector over Harris | corner detection · 'Lab' image | localization error (pixels): 1.471 -> 1.449 | Improved |
| increase anisotropic factor to rho^2=1.5 (K=8) (ablation) | SOGGDD detector · noise repeatability, 30 images | average repeatability | Hurt |
| use isotropic filters (rho^2=1) with K=4, 8, 16 directions (ablation) | SOGGDD detector · affine/JPEG/noise repeatability, 30 images | average repeatability | Improved |
| use K=8 directions with sigma^2=1.5 and rho^2=1 over other K values (ablation) | SOGGDD detector · repeatability, 30 images | average repeatability | Improved |
| use multi-direction second-order filtering rather than horizontal/vertical derivatives only | corner detection · Star-polygons 30 sides | missed and false corners | Improved |
| use SOGGDD detector on region repeatability | corner detection · Oxford sequences Trees, Bikes, Ubc, Leuven, Wall | repeatability score | Improved |
| use SOGGDD detector on Graffiti | corner detection · Oxford Graffiti sequence | repeatability score | No clear effect |
| use SOGGDD detector over ACJ in speed | corner detection · execution time, Bikes | execution time (s): 9.452 -> 4.521 | Improved |
| use SOGGDD detector over Harris in speed | corner detection · execution time, Bikes | execution time (s): 0.717 -> 4.521 | Hurt |
Closest papers by meaning
- Image Intensity Variation Information for Interest Point Detection (TPAMI 2023 · similarity 0.84)
- On Boundary Discontinuity in Angle Regression Based Arbitrary Oriented Object Detection (TPAMI 2024 · similarity 0.83)
- GauCho: Gaussian Distributions with Cholesky Decomposition for Oriented Object Detection (CVPR 2025 · similarity 0.83)
- Rethinking Boundary Discontinuity Problem for Oriented Object Detection (CVPR 2024 · similarity 0.81)
- ReDiffDet: Rotation-equivariant Diffusion Model for Oriented Object Detection (CVPR 2025 · similarity 0.81)
- Large Images Are Gaussians: High-Quality Large Image Representation with Levels of 2D Gaussian Splatting (AAAI 2025 · similarity 0.81)
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.
