Research map / Video / Video segmentation and tracking
XVFI: eXtreme Video Frame Interpolation
Hyeonjun Sim, Jihyong Oh, Munchurl Kim
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 |
|---|---|---|---|
| use complementary flow reversal instead of linear combination (ablation) | XVFI-Net · X-TEST | PSNR: 28.73 -> 28.86 | Improved |
| use complementary flow reversal instead of flow reversal (ablation) | XVFI-Net · X-TEST | PSNR: 28.30 -> 28.86 | Improved |
| use complementary flow reversal instead of linear combination (ablation) | XVFI-Net · X-TEST | tOF: 2.89 -> 2.67 | Improved |
| use complementary flow reversal instead of linear combination (ablation) | flow approximation · Vimeo90K test | EPE: 0.0752 -> 0.0721 | Improved |
| plain flow reversal is worse than linear approximation (ablation) | flow approximation · Vimeo90K test | EPE: 0.0752 -> 0.0892 | Hurt |
| train with S_trn=3 instead of 1, test at S_tst=3 (ablation) | XVFI-Net · X-TEST | PSNR: 28.40 -> 29.22 | Improved |
| test with deeper scale S_tst=5 instead of 3 (S_trn=3) (ablation) | XVFI-Net · X-TEST | PSNR: 29.22 -> 30.35 | Improved |
| test at S_tst=5 when trained with S_trn=1 (ablation) | XVFI-Net · X-TEST | PSNR: 28.40 -> 27.14 | Hurt |
| mismatch: train S_trn=5 but test at S_tst=1 (ablation) | XVFI-Net · X-TEST | PSNR: 29.48 -> 22.37 | Hurt |
| retrain baselines on X-TRAIN (ablation) | DAIN · X-TEST | PSNR: 26.78 -> 27.52 | Improved |
| retrain baselines on X-TRAIN (ablation) | AdaCoF · X-TEST | PSNR: 23.90 -> 25.81 | Improved |
| raise S_tst from 3 to 5 at inference (ablation) | XVFI-Net · X-TEST | PSNR: 28.86 -> 30.12 | Improved |
Closest papers by meaning
- BiM-VFI: Bidirectional Motion Field-Guided Frame Interpolation for Video with Non-uniform Motions (CVPR 2025 · similarity 0.89)
- Sparse Global Matching for Video Frame Interpolation with Large Motion (CVPR 2024 · similarity 0.89)
- IQ-VFI: Implicit Quadratic Motion Estimation for Video Frame Interpolation (CVPR 2024 · similarity 0.89)
- Disentangled Motion Modeling for Video Frame Interpolation (AAAI 2025 · similarity 0.88)
- Repurposing Pre-trained Video Diffusion Models for Event-based Video Interpolation (CVPR 2025 · similarity 0.88)
- Perception-Oriented Video Frame Interpolation via Asymmetric Blending (CVPR 2024 · similarity 0.88)
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.
