Efficient feedforward categorization of objects and human postures with address-event image sensors

This paper proposes an algorithm for feedforward categorization of objects and, in particular, human postures in real-time video sequences from address-event temporal-difference image sensors. The system employs an innovative combination of eventbased hardware and bio-inspired software architecture....

Descripción completa

Detalles Bibliográficos
Autores: Chen, Shoushun, Akselrod, Polina, Zhao, Bo, Pérez Carrasco, José Antonio, Linares Barranco, Bernabé, Culurciello, Eugenio
Tipo de recurso: artículo
Estado:Versión aceptada para publicación
Fecha de publicación:2012
País:España
Institución:Universidad de Sevilla (US)
Repositorio:idUS. Depósito de Investigación de la Universidad de Sevilla
OAI Identifier:oai:idus.us.es:11441/76597
Acceso en línea:https://hdl.handle.net/11441/76597
https://doi.org/10.1109/TPAMI.2011.120
Access Level:acceso abierto
Descripción
Sumario:This paper proposes an algorithm for feedforward categorization of objects and, in particular, human postures in real-time video sequences from address-event temporal-difference image sensors. The system employs an innovative combination of eventbased hardware and bio-inspired software architecture. An event-based temporal difference image sensor is used to provide input video sequences, while a software module extracts size and position invariant line features inspired by models of the primate visual cortex. The detected line features are organized into vectorial segments. After feature extraction, a modified line segment Hausdorffdistance classifier combined with on-the-fly cluster-based size and position invariant categorization. The system can achieve about 90 percent average success rate in the categorization of human postures, while using only a small number of training samples. Compared to state-of-the-art bio-inspired categorization methods, the proposed algorithm requires less hardware resource, reduces the computation complexity by at least five times, and is an ideal candidate for hardware implementation with event-based circuits.