AI Paper Atlas

Research map / Video / Video segmentation and tracking

Online Video Deblurring via Dynamic Temporal Blending Network

Tae Hyun Kim, Kyoung Mu Lee, Bernhard Scholkopf, Michael Hirsch

ICCV 2017 · Video › Video segmentation and tracking · approach: Spike and neuromorphic deblurring · 165 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
add spatio-temporal recurrence (STRCNN) to the plain CNN, 5 input frames (ablation)video deblurring CNN · 25 test videos, 5 framesPSNR (dB): 29.25 -> 29.63Improved
add dynamic temporal blending to STRCNN, 5 input frames (ablation)STRCNN · 25 test videos, 5 framesPSNR (dB): 29.63 -> 29.76Improved
add dynamic temporal blending to STRCNN, 3 input frames (ablation)STRCNN · 3 input framesPSNR (dB): 29.45 -> 29.61Improved
add dynamic temporal blending to STRCNN, 7 input frames (ablation)STRCNN · 7 input framesPSNR (dB): 29.37 -> 29.50Improved
add spatio-temporal recurrence to CNN across blur strengths (ablation)video deblurring CNN · varying blur strength, 5 framesaverage PSNR gain over baseline: 0.2dB higher than the baselineImproved
add recurrence and dynamic temporal blending across blur strengths (ablation)video deblurring CNN · varying blur strength, 5 framesaverage PSNR gain over baseline: 0.37dB against the baselineImproved
increase input blurry frames from 5 to 7 in STRCNN+DTB (ablation)STRCNN+DTB · 25 test videosPSNR (dB): 29.76 -> 29.50Hurt
reduce input frames from 5 to 3 in STRCNN+DTB (ablation)STRCNN+DTB · 25 test videosPSNR (dB): 29.76 -> 29.61Hurt
STRCNN+DTB vs Su et al. FLOW baselinevideo deblurring · 25 test videosPSNR (dB): 28.81 -> 29.02Improved
STRCNN+DTB online run vs Su et al. NOALIGNvideo deblurring · 100 HD framesrun time (sec): ~21 -> ~12.5Improved
place dynamic temporal blending between encoder and decoder rather than inside them (ablation)STRCNN+DTB · layout choicePSNRImproved

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