Research map / Image generation / Fast diffusion sampling
Efficient Image Retrieval via Decoupling Diffusion into Online and Offline Processing
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 |
|---|---|---|---|
| Use proposed diffusion instead of Iscen's diffusion | R-MAC global features (VGG) · Oxford105k | mAP: 82.7 -> 86.8 | Helped |
| Use proposed diffusion instead of Iscen's diffusion | R-MAC global features (ResNet) · Oxford105k | mAP: 87.4 -> 91.8 | Helped |
| Use proposed diffusion instead of Iscen's diffusion | R-MAC global features (VGG) · Paris106k | mAP: 92.5 -> 92.9 | Helped |
| Use proposed diffusion instead of Iscen's diffusion | R-MAC global features (ResNet) · Paris106k | mAP: 95.4 -> 95.6 | Helped |
| Fuse global and regional diffusion scores (ablation) | Proposed diffusion with R-MAC (VGG) · Oxford105k | mAP: 88.6 -> 91.2 | Helped |
| Fuse global and regional diffusion scores (ablation) | Proposed diffusion with R-MAC (ResNet) · Paris106k | mAP: 95.6 -> 96.2 | Helped |
| Apply truncation after full-graph normalization rather than before normalization (ablation) | Diffusion on Oxford datasets · Oxford datasets | mAP | Helped |
| Apply truncation after full-graph normalization rather than before normalization (ablation) | Diffusion on Paris datasets · Paris datasets | mAP | No clear effect |
| Use regional and global features with late fusion rather than regional features alone (ablation) | Proposed diffusion with R-MAC (VGG) · Paris6k | mAP: 93.9 -> 96.1 | Hurt |
Closest papers by meaning
- A Simple Early Exiting Framework for Accelerated Sampling in Diffusion Models (ICML 2024 · similarity 0.84)
- Cluster-Aware Similarity Diffusion for Instance Retrieval (ICML 2024 · similarity 0.84)
- Attention-Driven Training-Free Efficiency Enhancement of Diffusion Models (CVPR 2024 · similarity 0.84)
- Simpler Diffusion: 1.5 FID on ImageNet512 with Pixel-space Diffusion (CVPR 2025 · similarity 0.83)
- Where's Waldo: Diffusion Features For Personalized Segmentation and Retrieval (NeurIPS 2024 · similarity 0.83)
- Fast Direct: Query-Efficient Online Black-box Guidance for Diffusion-model Target Generation (ICLR 2025 · 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.