AggreGaze: Collective Estimation of Audience Attention on Public Displays

on YouTube ↗19m

" summary

AggreGaze presents a way to estimate where crowds look on public displays using only a single off‑the‑shelf camera mounted to the display. It corrects errors from appearance‑based gaze estimators by collecting on‑site training data (using stimuli like 9‑point calibration, moving targets, embedded visual elements, or regular video) and then aggregates many uncalibrated individual gaze estimates into spatio‑temporal attention maps that match ground‑truth distributions in a two‑week public deployment.

AI-written summary · about

" related talks

• 19mEveryday Eye Contact Detection Using Unsupervised Gaze Target DiscoveryAndreas Bulling, Yusuke Sugano, Xucong Zhang · UIST 2017
• 19mGazeProjector: Accurate Gaze Estimation and Seamless Gaze Interaction Across ...Sebastian Boring, Andreas Bulling, Christian Lander, Sven Gehring, Antonio Krger · UIST 2015
• 09mGazeChat: Enhancing Virtual Conferences with Gaze Awareness and Interactive 3D PhotosRuofei Du, Ken Perlin, Zhenyi He, Keru Wang, Brandon Yushan Feng · UIST 2021
• 20mBuilding a Personalized, Auto-Calibrating Eye Tracker from User InteractionsMichael Xuelin Huang, Tiffany C.K. Kwok, Grace Ngai, Stephen C.F. Chan, Hong Va Leong · CHI 2016
• 17mEyeGrip: Detecting Targets in a Series of Uni-directional Moving Objects Using Optokinetic ...Shahram Jalaliniya, Diako Mardanbegi · CHI 2016
• 21mSelf-Calibrating Head-Mounted Eye Trackers Using Egocentric Visual SaliencyAndreas Bulling, Yusuke Sugano · UIST 2015
• 19mEyeScout: Active Eye Tracking for Position and Movement Independent Gaze Interaction with Large ...Florian Alt, Mohamed Khamis, Andreas Bulling, Axel Hoesl, Alexander Klimczak, Martin Reiss · UIST 2017
• 15mCan Eye Help You?: Effects of Visualizing Eye Fixations on Remote Collaboration Scenarios for ...Yoichi Sato, Keita Higuch, Ryo Yonetani · CHI 2016