AI Paper Atlas

Research map / Video / Video segmentation and tracking

Spike-Temporal Latent Representation for Energy-Efficient Event-to-Video Reconstruction

Jianxiong Tang, Jian-Huang Lai, Lingxiao Yang, Xiaohua Xie

ECCV 2024 · Video › Video segmentation and tracking · approach: Neural video representations · 2 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 STLR spiking latent coding instead of EVSNNSNN-based E2V reconstruction · IJRRMSE: 0.097 -> 0.075Improved
use STLR instead of EVSNN for energy costSNN-based E2V reconstruction · IJRRenergy (J): 2.28 × 10−2 -> 1.26 × 10−3Improved
STLR compared to ANN E2VID on IJRR MSESTLR · IJRRMSE: 0.064 -> 0.075Hurt
add fixed point approximation (FPA) loss (ablation)STLR · Average over IJRR/HQF/MVSECMSE: 0.148 -> 0.104Improved
add FPA loss on HQF (ablation)STLR · HQFLPIPS: 0.400 -> 0.408Mixed
increase SVT unfolding steps from 5 to 10 (ablation)STLR · Average over IJRR/HQF/MVSECMSE: 0.104 -> 0.089Improved
increase SVT unfolding steps from 5 to 8 (ablation)STLR · Average over IJRR/HQF/MVSECMSE: 0.104 -> 0.117Hurt
choose decoder spiking neuron (LIF, PLIF, SPSN, SLTT) (ablation)STLR decoder · Average over IJRR/HQF/MVSECMSE: 0.111 -> 0.104No clear effect
inject Poisson noise (lambda=0.1) into input eventsSTLR vs E2VID · IJRRLPIPS: 0.566 -> 0.124Improved

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