Research map / Video / Video segmentation and tracking
Softmax Splatting for Video Frame Interpolation
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 softmax splatting instead of summation splatting (ablation) | Frame interpolation · Middlebury and Vimeo-90k test | PSNR | Helped |
| Use average splatting instead of linear splatting (ablation) | Frame interpolation · Middlebury and Vimeo-90k test | PSNR | Mixed |
| Refine the brightness-constancy importance metric with a small U-Net (ablation) | Softmax Splatting frame interpolation · Frame interpolation evaluation | PSNR | Helped |
| Add feature levels to the feature pyramid (ablation) | Softmax Splatting frame interpolation · Frame interpolation evaluation | interpolation quality | Helped |
| Use four feature levels instead of three (ablation) | Softmax Splatting frame interpolation · Vimeo-90k | interpolation quality | Hurt |
| Fine-tune PWC-Net for frame interpolation (ablation) | Softmax Splatting frame interpolation · Frame interpolation evaluation | interpolation quality | Helped |
| Change training objective from color loss to perceptual loss (ablation) | Softmax Splatting frame interpolation · Frame interpolation evaluation | PSNR and SSIM | Hurt |
| Change training objective from color loss to perceptual loss (ablation) | Softmax Splatting frame interpolation · Frame interpolation evaluation | LPIPS | Helped |
| Use FlowNet2 or LiteFlowNet instead of PWC-Net (ablation) | Softmax Splatting frame interpolation · Frame interpolation evaluation | interpolation quality | No clear effect |
| Interpolate across arbitrary temporal positions using forward warping | Softmax Splatting frame interpolation · High frame-rate Sintel | temporal consistency | Helped |
Closest papers by meaning
- ViBiDSampler: Enhancing Video Interpolation Using Bidirectional Diffusion Sampler (ICLR 2025 · similarity 0.87)
- HiCoM: Hierarchical Coherent Motion for Dynamic Streamable Scenes with 3D Gaussian Splatting (NeurIPS 2024 · similarity 0.87)
- Learning Efficient Fuse-and-Refine for Feed-Forward 3D Gaussian Splatting (NeurIPS 2025 · similarity 0.86)
- Taming Video Diffusion Prior with Scene-Grounding Guidance for 3D Gaussian Splatting from Sparse Inputs (CVPR 2025 · similarity 0.86)
- Generative Inbetweening: Adapting Image-to-Video Models for Keyframe Interpolation (ICLR 2025 · similarity 0.86)
- High-Resolution Frame Interpolation with Patch-based Cascaded Diffusion (AAAI 2025 · 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.