Research map / Video / Video-language understanding
Query-Focused Video Summarization: Dataset, Evaluation, and a Memory Network Based Approach
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 |
|---|---|---|---|
| condition the sequential DPP kernel on the query using a memory network | SeqDPP · UTE videos (average) | F1: 30.92 -> 44.19 | Helped |
| use the memory-network sequential DPP summarizer instead of SH-DPP | query-focused video summarization · UTE videos (average) | F1: 33.38 -> 44.19 | Helped |
| use the memory-network sequential DPP summarizer instead of SubMod | generic video summarization · UTE videos with no query-relevant shots (average) | F1: 50.22 -> 50.29 | Helped |
| use the memory-network sequential DPP summarizer instead of Quasi | generic video summarization · UTE videos with no query-relevant shots (average) | F1: 44.77 -> 50.29 | Helped |
| add query-conditioned attention instead of uniform attention (ablation) | Memory-network sequential DPP · UTE videos (average) | F1: 37.73 -> 44.19 | Helped |
| add embedding matrix D to compute DPP kernels (ablation) | Memory-network sequential DPP · UTE videos (average) | F1: 38.81 -> 44.19 | Helped |
| change embedding size from 128D to 256D (ablation) | Memory-network sequential DPP · UTE videos (average) | F1: 44.19 -> 42.19 | Hurt |
| randomly remove shots from user summaries | concept-annotation evaluation metric · user summaries compared with corrupted summaries | recall | Hurt |
Closest papers by meaning
- Video Summarization via Semantic Attended Networks (AAAI 2018 · similarity 0.92)
- Video Summarization with Large Language Models (CVPR 2025 · similarity 0.91)
- Scaling Up Video Summarization Pretraining with Large Language Models (CVPR 2024 · similarity 0.88)
- Video Summarization Using Denoising Diffusion Probabilistic Model (AAAI 2025 · similarity 0.88)
- MMSum: A Dataset for Multimodal Summarization and Thumbnail Generation of Videos (CVPR 2024 · similarity 0.87)
- V2Xum-LLM: Cross-Modal Video Summarization with Temporal Prompt Instruction Tuning (AAAI 2025 · similarity 0.87)
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.