Skeletons and Silhouettes: Comparing User Representations at a Gesture-based Large Display
summary
A night-facing large projected display behind glass uses a Connect depth sensor to recognize swipe and arm-up gestures; deployed 56 days, logging 190 sessions and 483 detected people. The silhouette attracts more passers-by and yields about 70% serious browsing sessions with median stay ~70s; the skeleton representation produces more playful behavior and longer stays (~100s). Representation choice shifts engagement style rather than visibility.
AI-written summary · about
chapters (5)
- 0:00Introduction and motivation
- 2:14User representation importance
- 4:20Study design and hypotheses
- 6:08Results and findings
- 8:58Conclusion and Q&A
related talks
19mAn Eye For Design: Gaze Visualizations for Remote Collaborative Work
08mSubtle and Personal Workspace Requirements for Visual Search Tasks on Public Displays
19mInvestigating the Use of Spatial Interaction for 3D Data Visualization on Mobile Devices
44mEffects and Causes
20mPhone vs. Tangible in Museums: A Comparative Study
14mExpanding the Bounds of Seated Virtual Workspace
25mAlternative Smartphone Interactions based on Sensor Data
12mHow do you deliver a feature on the biggest stage in the world?