AI Paper Atlas

Research map / Video / Video segmentation and tracking

PhaseNet for Video Frame Interpolation

Simone Meyer, Abdelaziz Djelouah, Brian McWilliams, Alexander Sorkine-Hornung, Markus Gross, Christopher Schroers

CVPR 2018 · Video › Video segmentation and tracking · approach: Video frame interpolation · 202 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
Use PhaseNet instead of SepConvVideo frame interpolation · Challenging scenes with motion blur or lighting changesvisual interpolation qualityMixed
Use PhaseNet instead of PhaseVideo frame interpolation · Large-motion and high-frequency scenesvisual interpolation qualityImproved
Use PhaseNet instead of MDP-Flow2Video frame interpolation · Scenes with lighting changesvisual interpolation qualityImproved
Add phase loss to image loss (ablation)PhaseNet · Video frame interpolationsharpnessImproved
Train with image loss only, removing phase loss (ablation)PhaseNet · Video frame interpolationsharpnessHurt
Train with phase loss only, removing image loss (ablation)PhaseNet · Video frame interpolationinterpolation qualityHurt
Vary the phase-loss weighting factor ν within [0.1, 1] (ablation)PhaseNet · Video frame interpolationinterpolation quality: ν ∈ [0.1, 1]No clear effect
Reuse trained highest-level weights for additional pyramid levels instead of averaging lower-level phase values (ablation)PhaseNet · High-resolution interpolationmotion rangeImproved
Average lower-level phase values instead of reusing trained weights (ablation)PhaseNet · High-resolution interpolationmotion rangeHurt
Use the hierarchical training procedure (ablation)PhaseNet · Video frame interpolationtraining stabilityImproved

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