Research map / Video / Video segmentation and tracking
FeatureFlow: Robust Video Interpolation via Structure-to-Texture Generation
Check 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 |
|---|---|---|---|
| Use structure-to-texture feature generation for frame interpolation instead of optical-flow pixel synthesis | FeFlow · Vimeo90K test set | PSNR: 34.71 -> 35.28 | Helped |
| Use structure-to-texture feature generation for frame interpolation instead of optical-flow pixel synthesis | FeFlow · Adobe240fps | PSNR: 32.51 -> 32.66 | Helped |
| Use structure-to-texture feature generation for frame interpolation instead of optical-flow pixel synthesis | FeFlow · Occ | PSNR: 36.98 -> 37.12 | Helped |
| Add edge-channel structure guidance and edge loss (ablation) | FeFlow · Vimeo90K test set | PSNR: 34.16 -> 34.84 | Helped |
| Add edge-channel structure guidance and edge loss (ablation) | FeFlow · Adobe240fps | PSNR: 32.52 -> 32.54 | No clear effect |
| Add triangle alignment loss (ablation) | FeFlow · Vimeo90K test set | PSNR: 34.84 -> 35.28 | Helped |
| Add triangle alignment loss (ablation) | FeFlow · Adobe240fps | PSNR: 32.54 -> 32.66 | Helped |
| Use 16 flow-attention groups instead of 1 (ablation) | FeFlow · Occ | PSNR: 35.25 -> 37.12 | Helped |
| Use 16 flow-attention groups instead of 4 (ablation) | FeFlow · Vimeo90K test set | PSNR: 35.02 -> 35.28 | Helped |
| Use 16 flow-attention groups instead of 4 (ablation) | FeFlow · Adobe240fps | PSNR: 32.54 -> 32.66 | Helped |
Closest papers by meaning
- Depth-Aware Video Frame Interpolation (CVPR 2019 · similarity 0.90)
- ST-MFNet: A Spatio-Temporal Multi-Flow Network for Frame Interpolation (CVPR 2022 · similarity 0.90)
- OCAI: Improving Optical Flow Estimation by Occlusion and Consistency Aware Interpolation (CVPR 2024 · similarity 0.87)
- Perception-Oriented Video Frame Interpolation via Asymmetric Blending (CVPR 2024 · similarity 0.87)
- Disentangled Motion Modeling for Video Frame Interpolation (AAAI 2025 · similarity 0.86)
- Sparse Global Matching for Video Frame Interpolation with Large Motion (CVPR 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.