Research map / Video / Video segmentation and tracking
Aligning Anything: Hierarchical Motion Estimation for Video Frame Interpolation
Mengshun Hu, Zhihang Zhong, Yansheng Qiu, Zheng Wang, Xiao Sun
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 |
|---|---|---|---|
| add HCE (ablation) | IFRNet-M · Vimeo90K | PSNR: 35.70 -> 35.85 | Improved |
| add HML (ablation) | IFRNet-M · Vimeo90K | PSNR: 35.70 -> 35.87 | Improved |
| add HML on top of HCE (ablation) | IFRNet-M · Vimeo90K | PSNR: 35.85 -> 35.96 | Improved |
| add HFL on top of HCE and HML (ablation) | IFRNet-M · Vimeo90K | LPIPS: 0.0172 -> 0.0169 | Improved |
| add HFL on top of HML (ablation) | IFRNet-M · Vimeo90K | PSNR: 35.87 -> 35.84 | No clear effect |
| add object-level features to HCE (ablation) | IFRNet-M · Vimeo90K | PSNR: 35.76 -> 35.85 | Improved |
| use hierarchical motion loss instead of pixel-level only (ablation) | IFRNet-M · Vimeo90K | PSNR: 35.79 -> 35.87 | Improved |
| use object-level motion loss instead of pixel-level only (ablation) | IFRNet-M · Vimeo90K | PSNR: 35.79 -> 35.80 | No clear effect |
| use hierarchical feature loss instead of pixel-level only (ablation) | IFRNet-M · Vimeo90K | LPIPS: 0.0175 -> 0.0172 | Improved |
| add HCE module (ablation) | IFRNet-M · inference time 1280x720 | time (ms): 30 -> 38 | Hurt |
| plug in HCE, HML, HFL | AMT-G · SNU-FILM Extreme | PSNR: 25.38 -> 25.61 | Improved |
| plug in HCE, HML, HFL | AMT-G · Vimeo90K | PSNR: 36.42 -> 36.62 | Improved |
Closest papers by meaning
- Disentangled Motion Modeling for Video Frame Interpolation (AAAI 2025 · similarity 0.90)
- Sparse Global Matching for Video Frame Interpolation with Large Motion (CVPR 2024 · similarity 0.88)
- TimeTracker: Event-based Continuous Point Tracking for Video Frame Interpolation with Non-linear Motion (CVPR 2025 · similarity 0.88)
- Perception-Oriented Video Frame Interpolation via Asymmetric Blending (CVPR 2024 · similarity 0.88)
- BiM-VFI: Bidirectional Motion Field-Guided Frame Interpolation for Video with Non-uniform Motions (CVPR 2025 · similarity 0.87)
- IQ-VFI: Implicit Quadratic Motion Estimation for Video Frame Interpolation (CVPR 2024 · 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.
