Research map / Image generation / Autoregressive and latent generation
Active Image Indexing
Read the paperCheck 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 |
|---|---|---|---|
| Activate indexed images by optimizing image pixels toward their index quantization | IVF-PQ, SSCD ResNet50 · DISC21, Ref10k transformed queries (Avg.) | Recall@1: 0.45 -> 0.75 | Helped |
| Activate indexed images | IVF 1024 · DISC21, Query50k transformed queries (Avg.) | Recall@1: 0.47 -> 0.83 | Helped |
| Activate indexed images | IVF 1024 · DISC21, Query50k transformed queries | μAP: 0.16 -> 0.43 | Helped |
| Activate indexed images | OPQ 8x8 · DISC21, Query50k transformed queries | μAP: 0.48 -> 0.55 | Helped |
| Activate indexed images | PCA64 LSH · DISC21, Query50k transformed queries | μAP: 0.25 -> 0.39 | Helped |
| Use 1 optimization step at learning rate 10 instead of 10 steps at learning rate 1 (ablation) | Active indexing with IVF-PQ · DISC21, Ref10k transformed queries (Avg.) | Recall@1: 0.75 -> 0.72 | Hurt |
| Use 1 optimization step at learning rate 10 instead of 10 steps at learning rate 1 (ablation) | Active indexing with IVF-PQ · DISC21, identity queries | Recall@1: 1.00 -> 0.99 | Hurt |
| Use expectation over transformations during activation (ablation) | Active indexing with IVF-PQ · DISC21, Ref10k transformed queries (Avg.) | Recall@1: 0.75 -> 0.76 | Helped |
| Use expectation over transformations during activation (ablation) | Active indexing with IVF-PQ · DISC21, rotation 90 queries | Recall@1: 0.50 -> 0.64 | Helped |
Closest papers by meaning
- A Benchmark and Asymmetrical-Similarity Learning for Practical Image Copy Detection (AAAI 2023 · similarity 0.86)
- Similarity Search for Efficient Active Learning and Search of Rare Concepts (AAAI 2022 · similarity 0.85)
- Deep Supervised Hashing for Fast Image Retrieval (CVPR 2016 · similarity 0.85)
- Efficient Large-Scale Similarity Search Using Matrix Factorization (CVPR 2016 · similarity 0.85)
- MagicLens: Self-Supervised Image Retrieval with Open-Ended Instructions (ICML 2024 · similarity 0.83)
- Exploiting Distribution Constraints for Scalable and Efficient Image Retrieval (ICLR 2025 · similarity 0.82)
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.