AI Paper Atlas

Research map / Video / Video segmentation and tracking

Learning Event-Driven Video Deblurring and Interpolation

Songnan Lin, Jiawei Zhang, Jinshan Pan, Zhe Jiang, Dongqing Zou, Yongtian Wang, Jing Chen, Jimmy Ren

ECCV 2020 · Video › Video segmentation and tracking · approach: Event-based video reconstruction · 152 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 addition of residuals with multiplication (physical model) (ablation)Ours · Blur-DVS synthetic (deblurring and interpolation)PSNR: 29.24 -> 29.65Improved
add dynamic filtering for spatially variant threshold (ablation)Ours · Blur-DVS synthetic (deblurring and interpolation)PSNR: 28.64 -> 29.65Improved
add previous-frame information in keyframe estimation (ablation)Ours · Blur-DVS synthetic (deblurring and interpolation)PSNR: 29.42 -> 29.65Improved
add adaptively-selected fusion (gate map) instead of 3D convs (ablation)Ours · Blur-DVS synthetic (deblurring and interpolation)PSNR: 29.34 -> 29.65Improved
add previous-frame information in keyframe estimation (ablation)Ours · Blur-DVS synthetic (deblurring and interpolation)SSIM: 0.855 -> 0.890Improved
add dynamic filtering (ablation)Ours · Blur-DVS synthetic (deblurring and interpolation)SSIM: 0.872 -> 0.890Improved
feed event data into STFANSTFAN · GoPro (video deblurring)PSNR: 30.28 -> 38.17Improved
feed event data into STFANSTFAN · Blur-DVS synthetic (video deblurring)PSNR: 19.03 -> 30.18Improved
feed intensity frames into event-only E2VE2V · Blur-DVS synthetic (video deblurring)PSNR: 16.89 -> 24.81Improved
feed events into TNTTTNTT · GoPro (deblurring and interpolation)PSNR: 32.47 -> 35.90Improved
use proposed physics-based network instead of STFAN*STFAN* · GoPro (video deblurring)PSNR: 38.17 -> 38.74Improved
use proposed network instead of TNTT*TNTT* · GoPro (deblurring and interpolation)PSNR: 35.90 -> 37.99Improved

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