AI Paper Atlas

Research map / Video / Video segmentation and tracking

The Spatio-Temporal Poisson Point Process: A Simple Model for the Alignment of Event Camera Data

Cheng Gu, Erik Learned-Miller, Daniel Sheldon, Guillermo Gallego, Pia Bideau

ICCV 2021 · Video › Video segmentation and tracking · approach: Event-based video reconstruction · 42 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 Poisson point-process likelihood instead of pairwise-entropy EMinevent alignment · DAVIS 240C shapes rotationangular velocity RMS error (deg/s): 42.22 -> 25.89Improved
use Poisson point-process likelihood instead of EMinevent alignment · DAVIS 240C boxes rotationangular velocity RMS error (deg/s): 7.06 -> 6.73Improved
use Poisson point-process likelihood instead of EMinevent alignment · DAVIS 240C poster rotationangular velocity RMS error (deg/s): 10.86 -> 10.37Improved
use Poisson point-process likelihood instead of EMinevent alignment · DAVIS 240C dynamic rotationangular velocity RMS error (deg/s): 5.39 -> 5.19Improved
use Poisson point-process likelihood instead of CMaxevent alignment · DAVIS 240C translation (avg)linear velocity RMS error: 0.43 -> 0.38Improved
use Poisson point-process likelihood instead of EMinevent alignment · DAVIS 240C translation (avg)linear velocity RMS error: 0.64 -> 0.38Improved
marginalize out the Poisson rate with a Gamma prior instead of per-pixel ML estimate (ablation)Poisson point-process loss · DAVIS 240C rotation (avg)average RMS error (deg/s): 12.1 -> 12.0No clear effect
use Poisson loss instead of CMax (accuracy versus speed)event alignment · DAVIS 240C poster rotation / loss timeRMS error and time: 13.45 -> 10.37 RMS; 1.2ms -> 3.1msMixed
reduce event batch size for low-texture scenes (ablation)Poisson loss · DAVIS 240C shapes rotationangular velocity RMS error (deg/s)Improved
use affine angular velocity model instead of constant velocity (ablation)Poisson loss · DAVIS 240C shapes rotationevent alignment qualityImproved

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