Research map / Video / Video segmentation and tracking
Context-Aware Synthesis for Video Frame Interpolation
Read the paperCheck 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 |
|---|---|---|---|
| Replace pixel-wise bidirectional blending with learned frame synthesis (ablation) | Context-aware synthesis · Middlebury interpolation benchmark (average) | PSNR: 34.71 -> 36.93 | Helped |
| Add warped contextual information to the synthesis network (ablation) | Frame synthesis network · Middlebury interpolation benchmark (average) | PSNR: 35.48 -> 36.93 | Helped |
| Use Laplacian loss instead of L1 color loss (ablation) | Context-aware synthesis · Middlebury interpolation benchmark (average) | PSNR: 36.71 -> 36.93 | Helped |
| Use Laplacian loss instead of feature loss (ablation) | Context-aware synthesis · Middlebury interpolation benchmark (average) | PSNR: 35.95 -> 36.93 | Helped |
| Use PWC-Net flow instead of SPyNet flow (ablation) | Context-aware synthesis · Middlebury interpolation benchmark (average) | PSNR: 35.78 -> 36.93 | Helped |
| Use flow-based pre-warping instead of synthesizing directly from input frames (ablation) | Context-aware synthesis · Middlebury interpolation benchmark (average) | PSNR: 34.98 -> 36.93 | Helped |
| Use PWC-Net flow instead of H.264 motion vectors (ablation) | Context-aware synthesis · Middlebury interpolation benchmark (average) | PSNR: 35.44 -> 36.93 | Helped |
| Use ResNet-18 conv1 context instead of VGG-19 conv1_2 context (ablation) | Context-aware synthesis · Middlebury interpolation benchmark (average) | PSNR: 36.77 -> 36.93 | Helped |
| Use ResNet-18 conv1 context instead of VGG-19 conv1_1 context (ablation) | Context-aware synthesis · Middlebury interpolation benchmark (average) | PSNR: 36.69 -> 36.93 | Helped |
Closest papers by meaning
- Depth-Aware Video Frame Interpolation (CVPR 2019 · similarity 0.94)
- Super SloMo: High Quality Estimation of Multiple Intermediate Frames for Video Interpolation (CVPR 2018 · similarity 0.91)
- OCAI: Improving Optical Flow Estimation by Occlusion and Consistency Aware Interpolation (CVPR 2024 · similarity 0.89)
- FlowVid: Taming Imperfect Optical Flows for Consistent Video-to-Video Synthesis (CVPR 2024 · similarity 0.88)
- Disentangled Motion Modeling for Video Frame Interpolation (AAAI 2025 · similarity 0.88)
- Hierarchical Flow Diffusion for Efficient 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.