2017/10/05 by Luiza Mici, Mici, Luiza, German I. Parisi +3
Computer Science · Neuroscience · #Anomaly Detection Techniques and Applications #Computer Vision and Pattern Recognition (cs.CV) #EEG and Brain-Computer Interfaces #FOS: Computer and information sciences #Human Pose and Action Recognition #Human-Computer Interaction (cs.HC) #Neural and Evolutionary Computing (cs.NE)
paper · pdf · doi:10.48550/arxiv.1710.01916
openalex publication_date 2017/10/05 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
The visual recognition of transitive actions comprising human-object\ninteractions is a key component for artificial systems operating in natural\nenvironments. This challenging task requires jointly the recognition of\narticulated body actions as well as the extraction of semantic elements from\nthe scene such as the identity of the manipulated objects. In this paper, we\npresent a self-organizing neural network for the recognition of human-object\ninteractions from RGB-D videos. Our model consists of a hierarchy of\nGrow-When-Required (GWR) networks that learn prototypical representations of\nbody motion patterns and objects, accounting for the development of\naction-object mappings in an unsupervised fashion. We report experimental\nresults on a dataset of daily activities collected for the purpose of this\nstudy as well as on a publicly available benchmark dataset. In line with\nneurophysiological studies, our self-organizing architecture exhibits higher\nneural activation for congruent action-object pairs learned during training\nsessions with respect to synthetically created incongruent ones. We show that\nour unsupervised model shows competitive classification results on the\nbenchmark dataset with respect to strictly supervised approaches.\n