AI Paper Atlas

Research map / Video / Video generation

Temporal Residual Guided Diffusion Framework for Event-Driven Video Reconstruction

Lin Zhu, Yunlong Zheng, Yijun Zhang, Xiao Wang, Lizhi Wang, Hua Huang

ECCV 2024 · Video › Video generation · approach: Efficient video diffusion models · 4 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 temporal residual as diffusion target instead of the intensity image (ablation)temporal residual guided diffusion · MVSECSSIM: 0.2565 -> 0.2768Improved
add cross-encoder attention (ablation)temporal residual guided diffusion · MVSECSSIM: 0.2653 -> 0.2768Improved
add event conditioning to the diffusion model (ablation)temporal residual guided diffusion · MVSECLPIPS: 0.3189 -> 0.2892Improved
accumulate temporal event features with ConvLSTM across time (ablation)temporal residual guided diffusion · MVSEC (qualitative and Fig. 6)brightness and contrastImproved
increase diffusion inference steps up to about 1,000 (ablation)temporal residual guided diffusion · IJRR, HQF, MVSECSSIM and LPIPSImproved
use the proposed method instead of E2VID+ on HQFevent video reconstruction · HQFMSE: 0.0363 -> 0.0441Hurt
use the proposed method versus the second-best on SSIMevent video reconstruction · IJRR, HQF, MVSECSSIM: improvements of 4.3%, 3.9% and 5.4% over the second-bestImproved

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