Research map / Video / Video generation
Towards Smooth Video Composition
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 |
|---|---|---|---|
| add alias-free generator operations (ablation) | StyleGAN-V · SkyTimelapse | FID: 40.8 -> 54.0 | Hurt |
| add image pretraining after adopting alias-free operations (ablation) | StyleGAN-V · YouTube Driving | FID: 56.4 -> 15.6 | Helped |
| add image pretraining after adopting alias-free operations (ablation) | StyleGAN-V · SkyTimelapse | FVD_16: 118.8 -> 73.5 | Helped |
| add TSM to the discriminator (ablation) | Config-C · SkyTimelapse | FID: 52.2 -> 49.9 | Hurt |
| add TSM to the discriminator (ablation) | Config-C · SkyTimelapse | FVD_16: 73.5 -> 49.0 | Helped |
| add TSM to the discriminator (ablation) | Config-C · YouTube Driving | FVD_128: 447.5 -> 221.5 | Helped |
| replace linear motion interpolation with B-Spline interpolation (ablation) | Config-D · Taichi-HD | FVD_16: 84.6 -> 82.6 | Helped |
| replace linear motion interpolation with B-Spline interpolation (ablation) | Config-D · YouTube Driving | FVD_16: 207.2 -> 281.6 | Hurt |
| add low-rank temporal modulation (ablation) | B-Spline motion representation · YouTube Driving | FVD_16: 281.6 -> 260.4 | Helped |
| add low-rank temporal modulation (ablation) | B-Spline motion representation · Taichi-HD | FVD_16: 82.6 -> 98.3 | Hurt |
| use B-Spline motion representation and low-rank temporal modulation (ablation) | Config-D · YouTube Driving | FVD_16-anchor: 269.1 -> 250.2 | Helped |
| use the short-video configuration rather than Config-F (ablation) | StyleSV · YouTube Driving long video | user preference (%): 12 -> 80 | Hurt |
Closest papers by meaning
- Generating Videos with Scene Dynamics (NeurIPS 2016 · similarity 0.90)
- Generating Videos with Dynamics-aware Implicit Generative Adversarial Networks (ICLR 2022 · similarity 0.89)
- Generating Long Videos of Dynamic Scenes (NeurIPS 2022 · similarity 0.89)
- Learning to Generate Time-Lapse Videos Using Multi-Stage Dynamic Generative Adversarial Networks (CVPR 2018 · similarity 0.89)
- Generative Inbetweening through Frame-wise Conditions-Driven Video Generation (CVPR 2025 · similarity 0.88)
- SF-V: Single Forward Video Generation Model (NeurIPS 2024 · similarity 0.86)
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.