Research map / Video / Video segmentation and tracking
TransVCL: Attention-Enhanced Video Copy Localization Network with Flexible Supervision
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 |
|---|---|---|---|
| Replace handcrafted cosine similarity with attention-enhanced features and end-to-end optimization (ablation) | TransVCL on VCSL · VCSL | F-score: 62.39 -> 66.51 | Helped |
| Add self-attention and cross-attention (ablation) | Basic localization model on VCSL · VCSL | F-score: 62.39 -> 64.74 | Helped |
| Add video-level classification loss with attention (ablation) | TransVCL on VCSL · VCSL | F-score: 64.74 -> 65.33 | Helped |
| Add temporal encoding with attention (ablation) | TransVCL on VCSL · VCSL | F-score: 64.74 -> 66.34 | Helped |
| Use the full attention-enhanced model instead of the basic localization model (ablation) | TransVCL · Video-level copy detection | F-score: 81.22 -> 90.19 | Helped |
| Add unlabeled training data | TransVCL with 1% labeled VCSL data · VCSL 1% labeled | F-score: 23.10 -> 37.28 (+14.18) | Helped |
| Add weakly labeled training data | TransVCL with 1% labeled VCSL data · VCSL 1% labeled | F-score: 23.10 -> 44.87 (+21.77) | Helped |
| Add unlabeled training data | TransVCL with 2% labeled VCSL data · VCSL 2% labeled | F-score: 31.95 -> 44.80 (+12.85) | Helped |
| Add weakly labeled training data | TransVCL with 2% labeled VCSL data · VCSL 2% labeled | F-score: 31.95 -> 50.50 (+18.55) | Helped |
| Add unlabeled training data | TransVCL with 5% labeled VCSL data · VCSL 5% labeled | F-score: 47.71 -> 56.11 (+ 8.40) | Helped |
| Add weakly labeled training data | TransVCL with 5% labeled VCSL data · VCSL 5% labeled | F-score: 47.71 -> 58.03 (+10.32) | Helped |
| Add unlabeled training data | TransVCL with 10% labeled VCSL data · VCSL 10% labeled | F-score: 59.07 -> 61.98 (+2.91) | Helped |
Closest papers by meaning
- Weakly-Supervised Video Re-Localization with Multiscale Attention Model (AAAI 2020 · similarity 0.88)
- Unsupervised Domain Adaptative Temporal Sentence Localization with Mutual Information Maximization (AAAI 2024 · similarity 0.84)
- SyncVIS: Synchronized Video Instance Segmentation (NeurIPS 2024 · similarity 0.84)
- Transferable Video Moment Localization by Moment-Guided Query Prompting (AAAI 2024 · similarity 0.84)
- Dense Audio-Visual Event Localization Under Cross-Modal Consistency and Multi-Temporal Granularity Collaboration (AAAI 2025 · similarity 0.83)
- HyperVAttention: Efficient Sparse Attention with Spatio-Temporal Clustering for Video Diffusion (NeurIPS 2026 · 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.