Research map / Video / Video-language understanding
Joint Commonsense and Relation Reasoning for Image and Video Captioning
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 |
|---|---|---|---|
| add commonsense reasoning (replace Faster R-CNN graph) (ablation) | C-R Reasoning captioner · MSVD test | CIDEr: 79.5 -> 96.8 | Helped |
| add commonsense reasoning (replace Faster R-CNN graph) (ablation) | C-R Reasoning captioner · MSVD test | BLEU-4: 48.6 -> 57.0 | Helped |
| add GCN relation reasoning (ablation) | C-R Reasoning captioner · MSVD test | CIDEr: 92.4 -> 96.8 | Helped |
| add GCN relation reasoning (ablation) | C-R Reasoning captioner · MSVD test | BLEU-4: 54.9 -> 57.0 | Helped |
| use C-R Reasoning instead of Up-Down (ablation) | C-R Reasoning captioner · MSVD test | CIDEr: 83.4 -> 96.8 | Helped |
| add Faster R-CNN detector to initial regions | C-R Reasoning captioner · MSCOCO Karpathy test | CIDEr: 117.3 -> 120.1 | Helped |
| add Faster R-CNN detector to initial regions | C-R Reasoning captioner · MSCOCO Karpathy test | SPICE: 20.1 -> 21.6 | Helped |
Closest papers by meaning
- Relational Distant Supervision for Image Captioning without Image-Text Pairs (AAAI 2024 · similarity 0.89)
- SGEITL: Scene Graph Enhanced Image-Text Learning for Visual Commonsense Reasoning (AAAI 2022 · similarity 0.89)
- Self-Critical Distillation Network for Video-based Commonsense Captioning (CVPR 2026 · similarity 0.88)
- Commonsense in Parts: Mining Part-Whole Relations from the Web and Image Tags (AAAI 2016 · similarity 0.88)
- Spatio-Temporal Graph for Video Captioning With Knowledge Distillation (CVPR 2020 · similarity 0.88)
- Composing Object Relations and Attributes for Image-Text Matching (CVPR 2024 · similarity 0.86)
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.