Research map / Video / Video-language understanding
Grounded Video Description
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 explicit box-based attention and classification supervision (ablation) | video description model · ActivityNet-Entities | grounding accuracy: 22.3 to 43.5 | Helped |
| add attention and region-classification supervision (ablation) | ActivityNet-Entities video description · ActivityNet-Entities val | BLEU@1: 23.2 to 23.9 | Helped |
| add attention and region-classification supervision (ablation) | ActivityNet-Entities video description · ActivityNet-Entities val | grounding accuracy: 21.3 to 41.6 | Helped |
| add self-attention to region encoding without grounding supervision (ablation) | unsupervised video description model · ActivityNet-Entities val | attention accuracy: 14.9 to 2.42 | Hurt |
| combine temporal attention with region attention (ablation) | grounded video description model · ActivityNet-Entities val | BLEU@4: 2.55, 2.45, 2.59 | Helped |
| use grounding supervision without classification supervision (ablation) | grounded video description model · ActivityNet-Entities val | classification accuracy: 14.9, 0.07 | Hurt |
| use classification supervision without attention or grounding supervision (ablation) | grounded video description model · ActivityNet-Entities val | classification accuracy: 6.06 to 14.9 | Helped |
| add box-based grounding supervision (ablation) | image description model · Flickr30k Entities test | grounding accuracy: 25.5 to 50.9 | Helped |
| add box-based grounding supervision (ablation) | image description model · Flickr30k Entities test | CIDEr: 60.1 to 62.3 | Helped |
| replace single attention module with temporal and region attention together (ablation) | grounded video description model · ActivityNet-Entities val | CIDEr: 43.5, 44.3, 47.5 | Helped |
| use Sup. Attn.+Cls. instead of the unsupervised baseline (ablation) | video description model · ActivityNet-Entities test | BLEU@4: 2.16 to 2.35 | Helped |
Closest papers by meaning
- Comprehensive Visual Grounding for Video Description (AAAI 2024 · similarity 0.91)
- VideoGrounding-DINO: Towards Open-Vocabulary Spatio-Temporal Video Grounding (CVPR 2024 · similarity 0.87)
- Exploiting Auxiliary Caption for Video Grounding (AAAI 2024 · similarity 0.87)
- Universal Video Temporal Grounding with Generative Multi-modal Large Language Models (NeurIPS 2025 · similarity 0.87)
- OmniViD: A Generative Framework for Universal Video Understanding (CVPR 2024 · similarity 0.87)
- Empower Words: DualGround for Structured Phrase and Sentence-Level Temporal Grounding (NeurIPS 2025 · 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.