Research map / Label-efficient and robust learning / Recognition and long-tail
Semantic Regularisation for Recurrent Image Annotation
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 |
|---|---|---|---|
| Add semantic regularisation at the CNN-RNN interface (ablation) | CNN-RNN image captioning · MS COCO validation split | CIDEr: 0.932 -> 1.054 | Helped |
| Add deep supervision to the CNN embedding (ablation) | CNN-RNN image captioning · MS COCO validation split | CIDEr: 0.932 -> 1.006 | Helped |
| Use the semantically explicit prediction layer instead of the penultimate feature layer as the RNN embedding (ablation) | Deeply supervised CNN-RNN · MS COCO validation split | CIDEr: 1.006 -> 1.054 | Helped |
| Add 1K noisy user tags as side information | S-CNN-RNN · NUS-WIDE | O-F1: 70.96 -> 77.15 | Helped |
| Fine-tune the CNN end-to-end rather than fixing its weights | S-CNN-RNN with 1K noisy tags · NUS-WIDE | O-F1: 76.77 -> 77.15 | Helped |
| Add 1K noisy user tags as side information | S-CNN-RNN · Microsoft COCO multi-label classification | O-F1: 72.47 -> 75.16 | Helped |
Closest papers by meaning
- CNN-RNN: A Unified Framework for Multi-Label Image Classification (CVPR 2016 · similarity 0.90)
- End-to-End Convolutional Semantic Embeddings (CVPR 2018 · similarity 0.88)
- Making Convolutional Networks Recurrent for Visual Sequence Learning (CVPR 2018 · similarity 0.88)
- Order-Free RNN With Visual Attention for Multi-Label Classification (AAAI 2018 · similarity 0.88)
- Pre-Trained Vision-Language Models as Noisy Partial Annotators (AAAI 2025 · similarity 0.85)
- Doomed to Re-Annotate, Forever: The ImageNet Story (NeurIPS 2026 · similarity 0.85)
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.