Research map / Detection and segmentation / Weakly supervised detection and segmentation
Kill Two Birds With One Stone: Weakly-Supervised Neural Network for Image Annotation and Tag Refinement
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 visual consistency alone rather than semantic dependency alone (ablation) | Weakly-supervised image annotation model · Mirflickr image annotation | mAP: 0.564 -> 0.572 | Helped |
| Use visual consistency alone rather than user-error sparsity alone (ablation) | Weakly-supervised image annotation model · Mirflickr image annotation | mAP: 0.560 -> 0.572 | Helped |
| Use visual consistency alone rather than semantic dependency alone (ablation) | Weakly-supervised tag refinement model · Mirflickr tag refinement | mAP: 0.615 -> 0.623 | Helped |
| Use visual consistency alone rather than user-error sparsity alone (ablation) | Weakly-supervised tag refinement model · Mirflickr tag refinement | mAP: 0.601 -> 0.623 | Helped |
| Use visual consistency alone rather than semantic dependency alone (ablation) | Weakly-supervised image annotation model · NUS-WIDE image annotation | mAP: 0.380 -> 0.386 | Helped |
| Use visual consistency alone rather than user-error sparsity alone (ablation) | Weakly-supervised image annotation model · NUS-WIDE image annotation | mAP: 0.375 -> 0.386 | Helped |
| Use visual consistency alone rather than semantic dependency alone (ablation) | Weakly-supervised tag refinement model · NUS-WIDE tag refinement | mAP: 0.425 -> 0.428 | Helped |
| Use visual consistency alone rather than user-error sparsity alone (ablation) | Weakly-supervised tag refinement model · NUS-WIDE tag refinement | mAP: 0.421 -> 0.428 | Helped |
| Combine visual consistency, semantic dependency, and user-error sparsity | Weakly-supervised model · Mirflickr and NUS-WIDE image annotation and tag refinement | mAP | Helped |
Closest papers by meaning
- Weakly-Supervised Deep Nonnegative Low-Rank Model for Social Image Tag Refinement and Assignment (AAAI 2017 · similarity 0.91)
- Tagging Like Humans: Diverse and Distinct Image Annotation (CVPR 2018 · similarity 0.87)
- Weakly Supervised Deep Image Hashing Through Tag Embeddings (CVPR 2019 · similarity 0.87)
- Revisiting Weakly Supervised Pre-Training of Visual Perception Models (CVPR 2022 · similarity 0.87)
- Doomed to Re-Annotate, Forever: The ImageNet Story (NeurIPS 2026 · similarity 0.84)
- Pre-Trained Vision-Language Models as Noisy Partial Annotators (AAAI 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.