AI Paper Atlas

Research map / Video / Video segmentation and tracking

EMoTive: Event-guided Trajectory Modeling for 3D Motion Estimation

Zengyu Wan, Wei Zhai, Yang Cao, Zhengjun Zha

ICCV 2025 · Video › Video segmentation and tracking · approach: Event-based video reconstruction · 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
add temporal gradient regularization Lt (ablation)EMoTive · CarlaEvent3Doptical flow EPE: 3.148 -> 3.118Improved
add temporal gradient regularization Lt (ablation)EMoTive · CarlaEvent3Dmotion in depth log-mid: 139.409 -> 138.585Improved
use event-guided non-uniform trajectory instead of uniform B-spline (ablation)EMoTive · CarlaEvent3Doptical flow EPE and log-mid: 4.6% (optical flow) and 6.9% (depth motion)Improved
use event-guided non-uniform trajectory instead of uniform B-spline (ablation)EMoTive · DSECoptical flow EPE: 0.497 -> 0.439Improved
increase temporal bins of Event Kymograph beyond 120 (ablation)EMoTive · CarlaEvent3DEPE / log-mid: optimal 120 bins / 100msHurt
use finer temporal resolution up to 120 bins (ablation)EMoTive · CarlaEvent3DEPE / log-mid: 60 to 120 binsImproved
raise curve degree beyond 3 or lower it (ablation)EMoTive · CarlaEvent3DEPE / log-mid: 3rd-order optimalMixed
set control points Nc=5 vs 4 (ablation)EMoTive · CarlaEvent3Doptical flow EPE: 0.024px decrease from Nc = 4 to 3.118pxImproved
set control points Nc=5 vs 6 (ablation)EMoTive · CarlaEvent3Dlog-mid: 1.6% improvement over Nc = 6 (138.585)Improved
use event-driven trajectory modeling instead of spatial-correlation methodsEMoTive vs image-based methods · CarlaEvent3Doptical flow EPE: 43.5% in optical flow EPE, and 33.6% in motion in depth log-midImproved
apply image-based spatial-correlation methods to event voxelsExpansion, TPCV, ScaleFlow, Scale++ · CarlaEvent3Doptical flow EPE: 5.274 - 1.963 EPE increaseHurt
EMoTive compared with ScaleFlow in size and speedEMoTive · CarlaEvent3Dparams / inference time: 5.61M vs 10.7M parameters; 40 ms, 52.9% speed improvementImproved

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