Research map / Video / Video generation
V-Trans4Style: Visual Transition Recommendation for Video Production Style Adaptation
Pooja Guhan, Tsung-Wei Huang, Guan-Ming Su, Subhadra Gopalakrishnan, Dinesh Manocha
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 |
|---|---|---|---|
| replace AutoTransition with the transformer encoder-decoder with masked triplet loss | transition recommendation · AutoTransition++ (pre-trained transition embedding) | Recall@5: 44.58 -> 56.06 | Improved |
| use pre-trained transition embeddings instead of random embeddings (ablation) | V-Trans4Style (E+D) · AutoTransition++ | Recall@5: 47.39 -> 56.06 | Improved |
| use MLN embeddings instead of AutoTransition's embeddings (ablation) | V-Trans4Style (E+D) · AutoTransition++ | Recall@1 / Recall@5: Recall@1 23.24 -> 22.59; Recall@5 56.06 -> 61.03 | Mixed |
| add masked triplet loss to classification loss (ablation) | V-Trans4Style (E+D) · AutoTransition++ | Recall@1: 11.63 -> 22.59 | Improved |
| add classification loss to masked triplet loss (ablation) | V-Trans4Style (E+D) · AutoTransition++ | Mean rank: 8.57 -> 5.90 | Improved |
| triplet loss alone versus classification loss alone (ablation) | V-Trans4Style (E+D) · AutoTransition++ | Recall@5: 69.00 -> 45.79 | Hurt |
| add style conditioning module without RRT (ablation) | V-Trans4Style (E&D) · Style cosine similarity, Vlog | cosine similarity: -0.01 -> 0.018 | Improved |
| add RRT fine-tuning of the transition sequence to SCM (ablation) | V-Trans4Style SCM · Style cosine similarity, Photos | cosine similarity: 0.0211 -> 0.7286 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- AutoTransition — AutoTransition: Learning to Recommend Video Transition Effects (ECCV 2022) · better on 2 of 3 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
- AutoTransition: Learning to Recommend Video Transition Effects (ECCV 2022 · similarity 0.91)
- StyleMaster: Stylize Your Video with Artistic Generation and Translation (CVPR 2025 · similarity 0.86)
- Semantix: An Energy-guided Sampler for Semantic Style Transfer (ICLR 2025 · similarity 0.85)
- VSTAR: Generative Temporal Nursing for Longer Dynamic Video Synthesis (ICLR 2025 · similarity 0.85)
- SEINE: Short-to-Long Video Diffusion Model for Generative Transition and Prediction (ICLR 2024 · similarity 0.85)
- Señorita-2M: A High-Quality Instruction-based Dataset for General Video Editing by Video Specialists (NeurIPS 2025 · similarity 0.84)
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.
