AI Paper Atlas

Research map / Video / Video segmentation and tracking

E-NeRV: Expedite Neural Video Representation with Disentangled Spatial-Temporal Context

Zizhang Li, Mengmeng Wang, Huaijin Pi, Kechun Xu, Jianbiao Mei, Yong Liu

ECCV 2022 · Video › Video segmentation and tracking · approach: Neural video representations · 121 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 spatial fusion function Phi at network start (ablation)E-NeRV · 8 videos averagePSNR: 37.89 -> 38.09Improved
add spatial-temporal fusion F_theta (transformer) (ablation)E-NeRV · 8 videos averagePSNR: 38.09 -> 38.45Improved
add temporal instance normalization branch (ablation)E-NeRV · 8 videos averagePSNR: 38.45 -> 38.79Improved
disentangle spatial/temporal context and reallocate saved parameters to conv channels (ablation)NeRV-L · 8 videos averagePSNR: 36.74 -> 37.89Improved
use transformer instead of MLP for fusion (ablation)E-NeRV · 8 videos averagePSNR: 38.54 -> 38.79Improved
use transformer instead of 3x3 conv for fusion (ablation)E-NeRV · 8 videos averagePSNR: 38.67 -> 38.79Improved
use disentangled design instead of NeRV-Split at 7.2MNeRV · 8 videos averagePSNR: 35.78 -> 36.74Improved
use disentangled design instead of NeRV-C_S at 5.8MNeRV · 8 videos averagePSNR: 33.72 -> 35.91Improved
replace NeRV with E-NeRV at similar sizeNeRV-L · Bunny videoPSNR: 39.63 -> 42.87Improved
replace NeRV with E-NeRV at similar sizeNeRV-L · Beauty videoPSNR: 36.06 -> 36.72Improved
replace NeRV with E-NeRV for denoisingNeRV · Video denoisingPSNR: 34.69 -> 36.23Improved
lower IN frequency b from 1.25 to 1.05 in E-NeRVE-NeRV · Bunny unseen framesPSNR: 29.32 -> 33.77Improved

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