International Journal of Advanced Robotic Systems · 2013 · 25 citations · 20 references
ComputingComputer ScienceVision ScienceComputer VisionInteractive MediaMachine VisionFace PerceptionFace RecognitionFacial ExpressionFacial PerceptionObject DetectionObject RecognitionAd Audience MeasurementBroadcast Audience MeasurementDemographic InferenceDigital BroadcastingInterpupillary DistanceMonocular LocalizationSignageVisual Signage
The paper develops an interactive, content‑aware, cost‑effective digital signage system that adapts content to individual observers. The system uses a monocular camera and real‑time computer‑vision algorithms—Viola‑Jones face detection, multi‑view Active Appearance Models for face registration, interpupillary‑distance distance estimation, and SVM classifiers for gender and age—to extract observer features and broadcast tailored content, and was tested in laboratory and field studies. The monocular localization module achieved accuracy comparable to a Kinect stereo system, the facial characterization module reached 95 % gender and 92 % age accuracy on FERET, and field studies confirmed the system’s real‑world applicability.
In this paper we present the development of an interactive, content-aware and cost-effective digital signage system. Using a monocular camera installed within the frame of a digital signage display, we employ real-time computer vision algorithms to extract temporal, spatial and demographic features of the observers, which are further used for observer-specific broadcasting of digital signage content. The number of observers is obtained by the Viola and Jones face detection algorithm, whilst facial images are registered using multi-view Active Appearance Models. The distance of the observers from the system is estimated from the interpupillary distance of registered faces. Demographic features, including gender and age group, are determined using SVM classifiers to achieve individual observer-specific selection and adaption of the digital signage broadcasting content. The developed system was evaluated at the laboratory study level and in a field study performed for audience measurement research. Comparison of our monocular localization module with the Kinect stereo-system reveals a comparable level of accuracy. The facial characterization module is evaluated on the FERET database with 95% accuracy for gender classification and 92% for age group. Finally, the field study demonstrates the applicability of the developed system in real-life environments.
20
Robust Real-Time Face Detection
Paul Viola, Michael Jones · International Journal of Computer Vision · 2004 · 14.2K citations
Active Appearance Models Revisited
Iain Matthews, Simon Baker · International Journal of Computer Vision · 2004 · 1.6K citations
Context‐Aware Recommender Systems
Gediminas Adomavičius, Bamshad Mobasher, Francesco Ricci⋆ et al. · AI Magazine · 2011 · 1.4K citations · Full text
Research design and statistical analysis
Choice Reviews Online · 2011 · 1.2K citations