Research map / Video / Video generation
Generating Long Videos of Dynamic Scenes
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 |
|---|---|---|---|
| Train the low-resolution generator on long video sequences (ablation) | Proposed method · Long-video generation | FVD | Helped |
| Use a generator with a large temporal receptive field (ablation) | Proposed method · Long-video generation | FVD | Helped |
| Use temporal lowpass filters with appropriately sized footprints (ablation) | Proposed method · Long-video generation | FVD | Helped |
| Condition super-resolution on real rather than generated low-resolution videos (ablation) | Proposed method · Super-resolution evaluation | FVD | Helped |
| Use the proposed method instead of StyleGAN-V | Video generation · Horseback riding and mountain biking | FVD128 | Helped |
| Use the proposed method instead of StyleGAN-V | Video generation · ACID | FVD128 | Hurt |
| Use the proposed method instead of StyleGAN-V | Video generation · SkyTimelapse | FVD128 | Hurt |
| Use the proposed method instead of MoCoGAN-HD, TATS, or DIGAN | Video generation · SkyTimelapse | FVD128 | Helped |
| Use the proposed method instead of StyleGAN-V | Video generation · Horseback riding, mountain biking, ACID, and SkyTimelapse | human motion realism preference | Helped |
| Use the proposed method instead of StyleGAN-V | Video generation · Horseback riding, ACID, and SkyTimelapse | rate of change in perceptual similarity | Helped |
| Use the proposed method instead of StyleGAN-V | Video generation · Mountain biking | rate of change in perceptual similarity | Mixed |
Closest papers by meaning
- WorldWeaver: Generating Long-Horizon Video Worlds via Rich Perception (NeurIPS 2025 · similarity 0.90)
- StreamingT2V: Consistent, Dynamic, and Extendable Long Video Generation from Text (CVPR 2025 · similarity 0.90)
- Generative Inbetweening through Frame-wise Conditions-Driven Video Generation (CVPR 2025 · similarity 0.89)
- FreeLong: Training-Free Long Video Generation with SpectralBlend Temporal Attention (NeurIPS 2024 · similarity 0.89)
- LongDiff: Training-Free Long Video Generation in One Go (CVPR 2025 · similarity 0.89)
- SV4D: Dynamic 3D Content Generation with Multi-Frame and Multi-View Consistency (ICLR 2025 · similarity 0.89)
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.