Research map / Learning theory and algorithms / Prediction and decision losses
Accumulated Stability Voting: A Robust Descriptor From Descriptors of Multiple Scales
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 |
|---|---|---|---|
| Replace SIFT with single-threshold ASV-SIFT | SIFT descriptor · Oxford | mAP: 0.3691 -> 0.4649 | Helped |
| Replace SIFT with single-threshold ASV-SIFT | SIFT descriptor · Fischer | mAP: 0.6219 -> 0.6623 | Helped |
| Add voting thresholding to raw accumulated stability (ablation) | AS-SIFT · Oxford | mAP: 0.4298 -> 0.4649 | Helped |
| Add voting thresholding to raw accumulated stability (ablation) | AS-SIFT · Fischer | mAP: 0.6447 -> 0.6623 | Helped |
| Use the alternative pP threshold selection scheme instead of median-based maximum-entropy thresholding (ablation) | ASV-SIFT · Oxford | mAP: 0.4649 -> 0.4479 | Hurt |
| Use the alternative pP threshold selection scheme instead of median-based maximum-entropy thresholding (ablation) | ASV-SIFT · Fischer | mAP: 0.6623 -> 0.6565 | Hurt |
| Add multiple first-stage thresholds (n1_t=3) instead of single thresholding (ablation) | ASV-SIFT · Oxford | mAP: 0.4649 -> 0.4731 | Helped |
| Increase first-stage threshold count from n1_t=3 to n1_t=15 (ablation) | ASV-SIFT · Oxford | mAP: 0.4731 -> 0.4740 | No clear effect |
| Use multiple thresholds at both stages rather than single thresholding at both stages (ablation) | Binary ASV-SIFT · Oxford | mAP: 0.3517 -> 0.4399 | Helped |
| Use multiple thresholds at both stages rather than single thresholding at both stages (ablation) | Binary ASV-SIFT · Fischer | mAP: 0.5855 -> 0.6483 | Helped |
| Add feature interpolation between neighboring scale samples (ablation) | ASV-SIFT · Oxford and Fischer | mAP | Helped |
| Increase the number of sampled scales (ablation) | ASV descriptors · Oxford and Fischer | mAP | Helped |
Closest papers by meaning
- Probabilistic Stability Guarantees for Feature Attributions (NeurIPS 2025 · similarity 0.84)
- U-MOF: Uncertainty-Guided Parameter-Efficient Multi-Objective Fine-Tuning for Long-Tailed Recognition (NeurIPS 2026 · similarity 0.82)
- Algorithmic stability and generalization of an unsupervised feature selection algorithm (NeurIPS 2021 · similarity 0.82)
- Selective Disk Bispectrum: A Complete and Rotation Invariant Image Descriptor (NeurIPS 2026 · similarity 0.82)
- DaWin: Training-free Dynamic Weight Interpolation for Robust Adaptation (ICLR 2025 · similarity 0.81)
- Robust Visual Recognition with Class-Imbalanced Open-World Noisy Data (AAAI 2024 · 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.