AI Paper Atlas

Research map / Video / Video segmentation and tracking

Video Frame Interpolation via Adaptive Separable Convolution

Simon Niklaus, Long Mai, Feng Liu

ICCV 2017 · Video › Video segmentation and tracking · approach: Video frame interpolation · 770 citations (OpenAlex)

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 changedWhereMeasuredEffect
replace L1-only training with L1 then perceptual (feature reconstruction) fine-tuning (ablation)Ours - L1 · Middlebury interpolation (average)interpolation error (all): 5.61 -> 5.81Hurt
replace L1-only training with L1 then perceptual (feature reconstruction) fine-tuningOurs - L1 · Cross-validation (250,000 samples)PSNR: 32.92 -> 32.37Hurt
replace L1-only training with L1 then perceptual (feature reconstruction) fine-tuningOurs - L1 · See You Again videoPSNR: 41.31 -> 40.88Hurt
add perceptual loss fine-tuning to the separable convolution network (ablation)Ours - L1 · visual quality (sharpness)user study / qualitativeImproved
estimate separable 1D kernels instead of 2D kernels (AdaConv)AdaConv · Middlebury interpolation (average)interpolation error (all): 6.20 -> 5.61Improved
estimate separable 1D kernels instead of 2D kernels (AdaConv)AdaConv · Cross-validation (250,000 samples)PSNR: 30.16 -> 32.92Improved
estimate separable 1D kernels instead of 2D kernels (AdaConv)AdaConv · 1080p interpolation speedspeed: over 20 times fasterImproved
use 1D separable kernels approximating 41x41 kernelsdense 2D kernel estimation · 1080p frame kernel memoryGB: 26 GB -> 1.27 GBImproved
use four sub-networks for the four kernels instead of a joint representation (ablation)kernel estimation network · trainingconvergence speedImproved
use bilinear upsampling in decoder instead of poorly chosen upsampling (ablation)decoder · interpolated framescheckerboard artifactsImproved
use L2 loss instead of L1 (ablation)separable conv network · interpolated framesvisual sharpnessHurt
synthesize frame directly instead of estimating kernels (ablation)separable conv network · interpolated framesvisual qualityHurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy