Research map / Video / Video segmentation and tracking
Training Weakly Supervised Video Frame Interpolation With Events
Zhiyang Yu, Yu Zhang, Deyuan Liu, Dongqing Zou, Xijun Chen, Yebin Liu, Jimmy S. Ren
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 AAFB adaptive fusion instead of concat+conv (ablation) | CAF · GoPro (CAF stage only) | PSNR: 32.27 -> 32.47 | Improved |
| add event branch to image-only CAF (ablation) | CAF · GoPro (CAF stage only) | PSNR: 29.43 -> 32.47 | Improved |
| add image branch to event-only CAF (ablation) | CAF · GoPro (CAF stage only) | PSNR: 31.37 -> 32.47 | Improved |
| use both image and event for keys and values (ablation) | SMT · GoPro (SMT ablation) | PSNR: 32.72 -> 33.33 | Improved |
| fuse first-stage appearance correction context in addition to second stage (ablation) | SMT · GoPro (SMT ablation) | PSNR: 32.56 -> 33.33 | Improved |
| fuse second-stage context in addition to first stage (ablation) | SMT · GoPro (SMT ablation) | PSNR: 33.00 -> 33.33 | Improved |
| use subpixel attention instead of hard attention (ablation) | SMT · GoPro (SMT ablation) | PSNR: 33.02 -> 33.33 | Improved |
| use subpixel attention instead of soft attention (ablation) | SMT · GoPro (SMT ablation) | PSNR: 32.50 -> 33.33 | Improved |
| train on low frame-rate triplets instead of high frame-rate videos | SloMo* · GoPro (10x interpolation) | PSNR: 32.79 -> 31.40 | Hurt |
| train on low frame-rate triplets instead of high frame-rate videos | QVI* · GoPro (10x interpolation) | PSNR: 33.07 -> 29.88 | Hurt |
| train on low frame-rate triplets instead of high frame-rate videos | QVI* · SloMo-DVS (4x interpolation) | PSNR: 33.70 -> 31.80 | Hurt |
| add events to the inputs of image-based interpolators | SloMo · GoPro (10x interpolation) | PSNR: 27.79 -> 32.79 | Improved |
Closest papers by meaning
- A Unified Framework for Event-Based Frame Interpolation With Ad-Hoc Deblurring in the Wild (TPAMI 2024 · similarity 0.90)
- Time Lens: Event-Based Video Frame Interpolation (CVPR 2021 · similarity 0.89)
- Repurposing Pre-trained Video Diffusion Models for Event-based Video Interpolation (CVPR 2025 · similarity 0.88)
- TimeTracker: Event-based Continuous Point Tracking for Video Frame Interpolation with Non-linear Motion (CVPR 2025 · similarity 0.86)
- Video Frame Prediction from a Single Image and Events (AAAI 2024 · similarity 0.86)
- TTA-EVF: Test-Time Adaptation for Event-based Video Frame Interpolation via Reliable Pixel and Sample Estimation (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.
