Research map / Point clouds / Point cloud classification
Quad-Networks: Unsupervised Learning to Rank for Interest Point Detection
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Learn a ranking detector from ground-truth correspondences instead of using DoG | Linear detector · Oxford VGG wall, 600 points | repeatability: 0.28 -> 0.36 | Helped |
| Learn a ranking detector from ground-truth correspondences instead of using DoG | Non-lin detector · Oxford VGG graf, 600 points | repeatability: 0.2 -> 0.19 | Hurt |
| Learn a ranking detector from ground-truth correspondences instead of using DoG | Linear detector · Oxford VGG bark, 600 points | repeatability: 0.13 -> 0.15 | Helped |
| Learn a ranking detector from ground-truth correspondences instead of using DoG | Linear detector · Oxford VGG boat, 600 points | repeatability: 0.25 -> 0.27 | Helped |
| Learn a ranking detector from ground-truth correspondences instead of using DoG | Non-lin detector · Oxford VGG leuven, 600 points | repeatability: 0.51 -> 0.72 | Helped |
| Learn a ranking detector from ground-truth correspondences instead of using DoG | Linear detector · Oxford VGG bikes, 600 points | repeatability: 0.41 -> 0.53 | Helped |
| Learn a ranking detector from ground-truth correspondences instead of using DoG | Non-lin detector · Oxford VGG trees, 600 points | repeatability: 0.3 -> 0.39 | Helped |
| Learn a ranking detector from ground-truth correspondences instead of using DoG | Linear detector · Oxford VGG ubc, 600 points | repeatability: 0.6 -> 0.62 | Helped |
| Train using small random warps rather than DoG | WarpS detector · Oxford VGG wall, 600 points | repeatability: 0.28 -> 0.32 | Helped |
| Train using small random warps rather than DoG | WarpS detector · Oxford VGG graf, 600 points | repeatability: 0.2 -> 0.17 | Hurt |
| Train using small random warps rather than DoG | WarpS detector · Oxford VGG bark, 600 points | repeatability: 0.13 -> 0.12 | Hurt |
| Train using small random warps rather than DoG | WarpS detector · Oxford VGG boat, 600 points | repeatability: 0.25 -> 0.21 | Hurt |
Closest papers by meaning
- Spatially Invariant Unsupervised Object Detection with Convolutional Neural Networks (AAAI 2019 · similarity 0.85)
- Unsupervised Representation Learning by Predicting Image Rotations (ICLR 2018 · similarity 0.84)
- Reinforced Feature Points: Optimizing Feature Detection and Description for a High-Level Task (CVPR 2020 · similarity 0.84)
- 4D Unsupervised Object Discovery (NeurIPS 2022 · similarity 0.83)
- Unsupervised Deep Learning by Neighbourhood Discovery (ICML 2019 · similarity 0.83)
- Unsupervised Image Matching and Object Discovery as Optimization (CVPR 2019 · 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.