Research map / Video / Video segmentation and tracking
Multiple Video Frame Interpolation via Enhanced Deformable Separable Convolution
Xianhang Cheng, Zhenzhong Chen
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 mask estimator (ablation) | EDSC · Vimeo90K | PSNR (dB): 34.62 -> 34.84 | Improved |
| add mask estimator (ablation) | EDSC · UCF101 | PSNR (dB): 35.00 -> 35.13 | Improved |
| add mask estimator (ablation) | EDSC · Middlebury Other | interpolation error: 2.13 -> 2.02 | Improved |
| add bias estimator (ablation) | EDSC · UCF101 | PSNR (dB): 35.03 -> 35.13 | Improved |
| add bias estimator (ablation) | EDSC · Vimeo90K | PSNR (dB): 34.80 -> 34.84 | Improved |
| increase kernel size from 1x1 to 5x5 (ablation) | EDSC · Vimeo90K | PSNR (dB): 33.47 -> 34.84 | Improved |
| increase kernel size from 3x3 to 5x5 (ablation) | EDSC · Vimeo90K | PSNR (dB): 34.59 -> 34.84 | Improved |
| increase kernel size beyond 5 (ablation) | EDSC · efficiency | FLOPs: 39.8%, 128.1% and 294.8% more FLOPs than n=5 | Hurt |
| reduce HetConv 3x3 rate from 1/1 to 1/4 (ablation) | EDSC encoder-decoder · Vimeo90K | PSNR (dB): 34.70 -> 34.84 | Improved |
| reduce HetConv 3x3 rate from 1/4 to 1/32 (ablation) | EDSC encoder-decoder · Vimeo90K | PSNR (dB): 34.84 -> 34.34 | Hurt |
| interpolate parameters toward L_F model (alpha 0 to 1) (ablation) | EDSC · Vimeo90K | PSNR (dB): 34.84 -> 34.49 | Hurt |
| interpolate parameters toward L_F model (alpha 0 to 1) (ablation) | EDSC · Vimeo90K | LPIPS: 0.026 -> 0.016 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- AdaCoF+ — AdaCoF: Adaptive Collaboration of Flows for Video Frame Interpolation (CVPR 2020) · better on 2 of 2 shared results
- MEMC-Net* — MEMC-Net: Motion Estimation and Motion Compensation Driven Neural Network for Video Interpolation and Enhancement (TPAMI 2019) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Video Frame Interpolation via Deformable Separable Convolution (AAAI 2020 · similarity 0.95)
- Video Frame Interpolation via Adaptive Separable Convolution (ICCV 2017 · similarity 0.93)
- Video Frame Interpolation via Adaptive Convolution (CVPR 2017 · similarity 0.90)
- Video Frame Interpolation without Temporal Priors (NeurIPS 2020 · similarity 0.88)
- Recurrent Partial Kernel Network for Efficient Optical Flow Estimation (AAAI 2024 · similarity 0.86)
- Generative Inbetweening: Adapting Image-to-Video Models for Keyframe Interpolation (ICLR 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.
