Research map / Video / Video-language understanding
Temporal Tessellation: A Unified Approach for Video Analysis
Dotan Kaufman, Gil Levi, Tal Hassner, Lior Wolf
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 |
|---|---|---|---|
| match clips in the joint semantics space instead of appearance space (ablation) | Local Tessellation · LSMDC16 | CIDEr-D: .042 -> .098 | Improved |
| add unsupervised Viterbi temporal coherence to local tessellation (ablation) | Tessellation · LSMDC16 | CIDEr-D: .098 -> .102 | Improved |
| add LSTM-predicted semantics dynamics (supervised) to local tessellation (ablation) | Tessellation · LSMDC16 | CIDEr-D: .098 -> .109 | Improved |
| add LSTM-predicted semantics dynamics to unsupervised tessellation (ablation) | Tessellation · LSMDC16 | BLEU-4: .007 -> .008 | Improved |
| add unsupervised temporal coherence to local tessellation (ablation) | Tessellation · SumMe | F-measure: 33.8 -> 41.4 | Improved |
| add supervised LSTM tessellation to local tessellation (ablation) | Tessellation · SumMe | F-measure: 33.8 -> 37.2 | Improved |
| use supervised instead of unsupervised tessellation (ablation) | Tessellation · SumMe | F-measure: 41.4 -> 37.2 | Hurt |
| add unsupervised temporal coherence to local tessellation (ablation) | Tessellation · TVSum | F-measure: 60.9 -> 64.1 | Improved |
| use supervised instead of unsupervised tessellation (ablation) | Tessellation · TVSum | F-measure: 64.1 -> 63.4 | Hurt |
| add context (supervised) to local tessellation (ablation) | Tessellation · THUMOS'14 IoT 0.1 | mAP: 56.4 -> 61.1 | Improved |
| add context (supervised) to local tessellation (ablation) | Tessellation · THUMOS'14 IoT 0.5 | mAP: 20.7 -> 23.3 | Improved |
| match in semantics space instead of appearance only (ablation) | Local Tessellation · Greatest Hits loudness | MSE error: 0.35 -> 0.27 | Improved |
Closest papers by meaning
- MMBench-Video: A Long-Form Multi-Shot Benchmark for Holistic Video Understanding (NeurIPS 2024 · similarity 0.89)
- MotionBench: Benchmarking and Improving Fine-grained Video Motion Understanding for Vision Language Models (CVPR 2025 · similarity 0.88)
- SF2T: Self-supervised Fragment Finetuning of Video-LLMs for Fine-Grained Understanding (CVPR 2025 · similarity 0.88)
- Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis (CVPR 2025 · similarity 0.88)
- MVBench: A Comprehensive Multi-modal Video Understanding Benchmark (CVPR 2024 · similarity 0.88)
- Vript: A Video Is Worth Thousands of Words (NeurIPS 2024 · 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.
