Research map / Vision-language and multimodal / Human and embodied modeling
Hierarchical Semantic Correspondence Networks for Video Paragraph Grounding
Read the paperCheck 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 hierarchical word-, sentence-, and paragraph-level correspondence and grounding | HSCNet · ActivityNet-Captions val 2 | mIoU: 55.91 -> 59.71 | Helped |
| use hierarchical word-, sentence-, and paragraph-level correspondence and grounding | HSCNet · TACoS test | mIoU: 31.42 -> 40.61 | Helped |
| add word-level and paragraph-level modeling to sentence-level grounding (ablation) | HSCNet · TACoS test | mIoU: 32.66 -> 40.61 | Helped |
| add word-level modeling to sentence- and paragraph-level modeling (ablation) | HSCNet · TACoS test | mIoU: 35.18 -> 40.61 | Helped |
| add paragraph-level modeling to sentence- and word-level modeling (ablation) | HSCNet · TACoS test | mIoU: 37.43 -> 40.61 | Helped |
| add word-level semantic alignment loss (ablation) | HSCNet encoder · TACoS test | mIoU: 22.38 -> 35.02 | Helped |
| include all word-, sentence-, and paragraph-level alignment losses rather than word-level loss alone (ablation) | HSCNet encoder · TACoS test | mIoU: 35.02 -> 40.61 | Helped |
| add paragraph-level decoder to sentence-level decoder (ablation) | HSCNet · TACoS test | mIoU: 37.57 -> 38.33 | Helped |
| add word-level decoder to sentence-level decoder (ablation) | HSCNet · TACoS test | mIoU: 37.57 -> 39.10 | Helped |
| add both word-level and paragraph-level decoders (ablation) | HSCNet · TACoS test | mIoU: 37.57 -> 40.61 | Helped |
Closest papers by meaning
- Learning Multi-Scale Video-Text Correspondence for Weakly Supervised Temporal Article Gronding (AAAI 2024 · similarity 0.92)
- Semi-Supervised Video Paragraph Grounding With Contrastive Encoder (CVPR 2022 · similarity 0.90)
- Siamese Learning with Joint Alignment and Regression for Weakly-Supervised Video Paragraph Grounding (CVPR 2024 · similarity 0.89)
- Empower Words: DualGround for Structured Phrase and Sentence-Level Temporal Grounding (NeurIPS 2025 · similarity 0.89)
- Local-Global Video-Text Interactions for Temporal Grounding (CVPR 2020 · similarity 0.88)
- Multi-Pair Temporal Sentence Grounding via Multi-Thread Knowledge Transfer Network (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.