Research map / Video / Video segmentation and tracking
Video Frame Interpolation Transformer
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 VFIT-S rather than FLAVR | video frame interpolation · Vimeo-90K test | PSNR: 36.30 -> 36.48 | Helped |
| use VFIT-B rather than FLAVR | video frame interpolation · DAVIS | PSNR: 27.44 -> 28.09 | Helped |
| replace patch-based global attention with local Sep-STS attention (ablation) | VFIT-B · Vimeo-90K ablation | PSNR: 35.18 -> 36.02 | Helped |
| use Sep-STS rather than inseparable STS blocks (ablation) | VFIT-B · Vimeo-90K ablation | PSNR: 35.84 -> 36.02 | Helped |
| use Sep-STS rather than convolutional ResBlocks (ablation) | VFIT-B · Vimeo-90K, fast motion | PSNR: 32.80 -> 33.23 | Helped |
| use Sep-STS rather than convolutional ResBlocks (ablation) | VFIT-B · Vimeo-90K, medium motion | PSNR: 35.75 -> 35.91 | Helped |
| use Sep-STS rather than convolutional ResBlocks (ablation) | VFIT-B · Vimeo-90K, slow motion | PSNR: 38.26 -> 38.36 | Helped |
| increase Sep-STS window size from M=4 to M=8 (ablation) | VFIT-B · Vimeo-90K ablation | PSNR: 35.82 -> 36.02 | Helped |
| increase Sep-STS window size from M=8 to M=10 (ablation) | VFIT-B · Vimeo-90K ablation | PSNR: 36.02 -> 35.93 | Hurt |
| use multi-scale synthesis instead of removing the second and third SynBlocks (ablation) | VFIT-B · Vimeo-90K test | PSNR: 35.54 -> 36.02 | Helped |
| supervise only the final output rather than all-scale outputs (ablation) | VFIT multi-scale synthesis · Vimeo-90K | PSNR | Helped |
| retain convolutional downsampling rather than use Transformer downsampling (ablation) | VFIT-B · Vimeo-90K ablation | PSNR: 35.92 -> 36.02 | Helped |
Closest papers by meaning
- Video Frame Interpolation With Transformer (CVPR 2022 · similarity 0.92)
- Video Frame Interpolation via Adaptive Convolution (CVPR 2017 · similarity 0.89)
- Channel Attention Is All You Need for Video Frame Interpolation (AAAI 2020 · similarity 0.89)
- Depth-Aware Video Frame Interpolation (CVPR 2019 · similarity 0.88)
- Super SloMo: High Quality Estimation of Multiple Intermediate Frames for Video Interpolation (CVPR 2018 · similarity 0.88)
- EDEN: Enhanced Diffusion for High-quality Large-motion Video Frame Interpolation (CVPR 2025 · 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.