Research map / Video / Video-language understanding
Localizing Moments in Video With Natural Language
Lisa Anne Hendricks, Oliver Wang, Eli Shechtman, Josef Sivic, Trevor Darrell, Bryan Russell
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 optical flow to RGB features (ablation) | LSTM-local · DiDeMo validation | Rank@1: 18.71 vs 13.10 (RGB only) | Improved |
| add RGB to flow features (fusion) (ablation) | LSTM-local · DiDeMo validation | Rank@1: 18.35 -> 18.71 | Improved |
| add global video context feature (ablation) | LSTM-Fusion-local · DiDeMo validation | Rank@1: 18.71 -> 19.88 | Improved |
| add global video context feature (ablation) | LSTM-Fusion-local · DiDeMo validation | Rank@5: 57.47 -> 62.39 | Improved |
| add temporal endpoint features (tef) (ablation) | LSTM-Fusion + global · DiDeMo validation | Rank@1: 19.88 -> 27.57 | Improved |
| add temporal endpoint features (tef) (ablation) | LSTM-Fusion + global · DiDeMo validation | Rank@5: 62.39 -> 79.69 | Improved |
| add temporal endpoint features (tef) (ablation) | LSTM-Fusion + global · DiDeMo validation | mIoU: 33.51 -> 41.70 | Improved |
| add intra-video negatives to inter-video negatives in ranking loss (ablation) | MCN · DiDeMo validation | Rank@1: 25.58 -> 27.57 | Improved |
| add inter-video negatives to intra-video negatives in ranking loss (ablation) | MCN · DiDeMo validation | Rank@1: 26.77 -> 27.57 | Improved |
| add inter-video negatives to intra-video negatives in ranking loss (ablation) | MCN · DiDeMo validation | mIoU: 39.83 -> 41.70 | Improved |
| add the text query to a context-only model (ablation) | MCN · DiDeMo validation | Rank@1 / Rank@5 / mIoU: 25.04, 75.23, 36.12 -> 27.57, 79.69, 41.70 | Improved |
| use MCN instead of moment frequency prior | MCN · DiDeMo test | Rank@1: 19.40 -> 28.10 | Improved |
Closest papers by meaning
- Multi-Scale 2D Temporal Adjacency Networks for Moment Localization With Natural Language (TPAMI 2021 · similarity 0.92)
- Learning 2D Temporal Adjacent Networks for Moment Localization with Natural Language (AAAI 2020 · similarity 0.91)
- Transferable Video Moment Localization by Moment-Guided Query Prompting (AAAI 2024 · similarity 0.91)
- Fewer Steps, Better Performance: Efficient Cross-Modal Clip Trimming for Video Moment Retrieval Using Language (AAAI 2024 · similarity 0.88)
- Boundary-Aware Temporal Dynamic Pseudo-Supervision Pairs Generation for Zero-Shot Natural Language Video Localization (AAAI 2025 · similarity 0.88)
- VERIFIED: A Video Corpus Moment Retrieval Benchmark for Fine-Grained Video Understanding (NeurIPS 2024 · similarity 0.88)
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.
