Research map / Video / Video-language understanding
Towards Generalisable Video Moment Retrieval: Visual-Dynamic Injection to Image-Text Pre-Training
Dezhao Luo, Jiabo Huang, Shaogang Gong, Hailin Jin, Yang Liu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 visual-dynamic injection to MMN with CLIP | MMN · Charades-STA novel-word OOD | R@1, IoU=0.5: 43.85 -> 46.47 | Improved |
| Add visual-dynamic injection to MMN with CLIP | MMN · ActivityNet-Captions novel-word OOD | R@1, IoU=0.7: 15.48 -> 16.02 | Improved |
| Add visual-dynamic injection to MMN with image-text pre-training | MMN · Charades-STA standard split | R@1, IoU=0.5: 50.48 -> 52.32 | Improved |
| Add visual-dynamic injection to MMN with image-text pre-training | MMN · ActivityNet-Captions standard split | R@1, IoU=0.5: 46.89 -> 48.09 | Improved |
| Use visual context injection without a text query (ablation) | MMN with CLIP on Charades-STA novel-word OOD · Charades-STA novel-word OOD | R@1, IoU=0.5: 43.85 -> 43.12 | Hurt |
| Use the complete query instead of the static query for visual context injection (ablation) | MMN with CLIP on Charades-STA novel-word OOD · Charades-STA novel-word OOD | R@1, IoU=0.5: 43.85 -> 45.02 | Improved |
| Use the static query to probe visual context (ablation) | MMN with CLIP on Charades-STA novel-word OOD · Charades-STA novel-word OOD | R@1, IoU=0.7: 24.17 -> 29.35 | Improved |
| Add spatial dynamic injection with an LSTM (ablation) | MMN with CLIP on Charades-STA novel-word OOD · Charades-STA novel-word OOD | R@1, IoU=0.5: 43.85 -> 45.32 | Improved |
| Add spatial dynamic injection with a Transformer encoder (ablation) | MMN with CLIP on Charades-STA novel-word OOD · Charades-STA novel-word OOD | R@1, IoU=0.5: 43.85 -> 44.60 | Improved |
| Combine visual context and spatial dynamic injection (ablation) | MMN with CLIP on Charades-STA novel-word OOD · Charades-STA novel-word OOD | R@1, IoU=0.5: 43.85 -> 46.47 | Improved |
Closest papers by meaning
- Generative Video Diffusion for Unseen Novel Semantic Video Moment Retrieval (AAAI 2025 · similarity 0.91)
- Towards Balanced Alignment: Modal-Enhanced Semantic Modeling for Video Moment Retrieval (AAAI 2024 · similarity 0.90)
- TOPA: Extending Large Language Models for Video Understanding via Text-Only Pre-Alignment (NeurIPS 2024 · similarity 0.90)
- VERIFIED: A Video Corpus Moment Retrieval Benchmark for Fine-Grained Video Understanding (NeurIPS 2024 · similarity 0.89)
- Zero-shot Video Moment Retrieval via Off-the-shelf Multimodal Large Language Models (AAAI 2025 · similarity 0.89)
- CDTR: Semantic Alignment for Video Moment Retrieval Using Concept Decomposition Transformer (AAAI 2025 · similarity 0.89)
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.
