Research map / Video / Video-language understanding
Object-Centric Framework for Video Moment Retrieval
Zongyao Li, Yongkang Wong, Satoshi Yamazaki, Jianquan Liu, Mohan Kankanhalli
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 object-centric framework to frame-level baseline (SF+C) (ablation) | SlowFast+CLIP-B frame-level baseline · Charades-STA | mIoU: 53.4 -> 56.2 | Improved |
| add object-centric framework to frame-level baseline (VideoMAEv2) (ablation) | VideoMAEv2-b frame-level baseline · Charades-STA | mIoU: 59.9 -> 61.3 | Improved |
| add object-centric framework to frame-level baseline (SF+C) (ablation) | SlowFast+CLIP-B frame-level baseline · QVHighlights val | mAP Avg.: 44.1 -> 49.9 | Improved |
| add object-centric framework to frame-level baseline (VideoMAEv2) (ablation) | VideoMAEv2-b frame-level baseline · QVHighlights val | mAP Avg.: 46.1 -> 50.8 | Improved |
| add both visual and semantic object features instead of only one (ablation) | object-centric framework · Charades-STA | mIoU: 55.7 vs 53.4 and 52.3 for single-feature variants | Improved |
| add frame-level branch (SF+C) to object-centric model (ablation) | object-centric framework · Charades-STA | mIoU: 55.7 -> 56.2 | Improved |
| add frame-level branch (VideoMAEv2) to object-centric model (ablation) | object-centric framework · Charades-STA | mIoU: 55.7 -> 61.3 | Improved |
| add frame-level branch (SF+C) to object-centric model (ablation) | object-centric framework · TACoS | mIoU: 43.1 -> 41.9 | Hurt |
| add frame-level branch (SF+C) to object-centric model (ablation) | object-centric framework · QVHighlights val | mAP Avg.: 47.1 -> 49.9 | Improved |
| use scene graph model trained on Action Genome instead of Visual Genome (ablation) | object-centric framework (no frame-level) · Charades-STA | mIoU: 55.7 -> 56.0 | No clear effect |
| use ground-truth scene graphs instead of Visual Genome predictions (ablation) | object-centric framework (no frame-level) · Charades-STA | mIoU: 55.7 -> 67.4 | Improved |
| use AG-trained scene graph model with VideoMAEv2 frame features (ablation) | object-centric framework + VideoMAEv2 · Charades-STA | mIoU: 61.3 -> 61.8 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- R2-Tuning — R^2-Tuning: Efficient Image-to-Video Transfer Learning for Video Temporal Grounding (ECCV 2024) · better on 2 of 3 shared results
- BAM-DETR — BAM-DETR: Boundary-Aligned Moment Detection Transformer for Temporal Sentence Grounding in Videos (ECCV 2024) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- The Devil is in the Spurious Correlations: Boosting Moment Retrieval with Dynamic Learning (ICCV 2025 · similarity 0.90)
- Transferable Video Moment Localization by Moment-Guided Query Prompting (AAAI 2024 · similarity 0.89)
- MomentSeeker: A Task-Oriented Benchmark For Long-Video Moment Retrieval (NeurIPS 2025 · similarity 0.89)
- Anchor-Aware Similarity Cohesion in Target Frames Enables Predicting Temporal Moment Boundaries in 2D (CVPR 2025 · similarity 0.89)
- When One Moment Isn't Enough: Multi-Moment Retrieval with Cross-Moment Interactions (NeurIPS 2025 · similarity 0.88)
- Generative Video Diffusion for Unseen Novel Semantic Video Moment Retrieval (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.
