Research map / Video / Video-language understanding
IntentVizor: Towards Generic Query Guided Interactive Video Summarization
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 |
|---|---|---|---|
| Use IntentVizor instead of the strongest listed baseline | Textual Query Dataset · Textual Query Dataset (Avg.) | F-1: 50.90 vs 49.20 | Helped |
| Use IntentVizor instead of the strongest listed baseline | Visual Query Dataset · Visual Query Dataset (Avg.) | F-1: 45.69 vs 38.55 | Helped |
| Use Ego-GCN rather than Transformer in both modules (ablation) | IntentVizor · Textual Query Dataset | F-1: 42.68 | Helped |
| Add Ego-GCN to the summary module while using Transformer in the intent module (ablation) | IntentVizor · Textual Query Dataset | F-1: 42.68 -> 46.15 | Helped |
| Add Ego-GCN to the intent module while using Transformer in the summary module (ablation) | IntentVizor · Textual Query Dataset | F-1: 42.68 -> 44.75 | Helped |
| Use Local GCN to recover shot-level features instead of upsampling (ablation) | IntentVizor · Textual Query Dataset | F-1: 35.88 -> 50.90 | Helped |
| Use Local GCN to recover shot-level features instead of transpose convolution (ablation) | IntentVizor · Textual Query Dataset | F-1: 45.18 -> 50.90 | Helped |
| Use both coarse- and fine-granularity pathways instead of shot-level features (ablation) | IntentVizor · Textual Query Dataset | F-1: 44.40 -> 50.90 | Helped |
| Add the fine-granularity pathway to the coarse-granularity pathway (ablation) | IntentVizor · Textual Query Dataset | F-1: 45.15 -> 50.90 | Helped |
| Add the coarse-granularity pathway to the fine-granularity pathway (ablation) | IntentVizor · Textual Query Dataset | F-1: 47.81 -> 50.90 | Helped |
| Fuse pathways early rather than at the middle stage (ablation) | IntentVizor · Textual Query Dataset | F-1: 46.66 | Helped |
| Fuse pathways early rather than at the late stage (ablation) | IntentVizor · Textual Query Dataset | F-1: 45.47 | Helped |
Closest papers by meaning
- Query-Focused Video Summarization: Dataset, Evaluation, and a Memory Network Based Approach (CVPR 2017 · similarity 0.91)
- V2Xum-LLM: Cross-Modal Video Summarization with Temporal Prompt Instruction Tuning (AAAI 2025 · similarity 0.87)
- Video Summarization with Large Language Models (CVPR 2025 · similarity 0.87)
- Scaling Up Video Summarization Pretraining with Large Language Models (CVPR 2024 · similarity 0.86)
- AdaVideoRAG: Omni-Contextual Adaptive Retrieval-Augmented Efficient Long Video Understanding (NeurIPS 2025 · similarity 0.83)
- Video-RAG: Visually-aligned Retrieval-Augmented Long Video Comprehension (NeurIPS 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.