作者: Edoardo Ardizzone , Marco La Cascia , Filippo Vella
DOI: 10.1007/978-3-540-76856-2_26
关键词: Computer vision 、 Artificial intelligence 、 Context (language use) 、 Face detection 、 Index (publishing) 、 Exploit 、 Computer science 、 Image retrieval 、 Pattern recognition (psychology) 、 Domain (software engineering) 、 Term (time)
摘要: The proposed work exploits methods and techniques for automatic characterization of images content-based access to personal photo libraries. Several techniques, even if not reliable enough address the general problem image retrieval, have been proven quite robust in a limited domain such as one album. In particular, starting from observation that most photos depict usually small number people relatively different contexts (e.g. Beach, Public Garden, Indoor, Nature, Snow, City, etc...) we propose use borrowed fields computer vision pattern recognition index based on who is present scene context where picture were taken. Experiments collection about thousand proved simple lead surprisingly good results term easyness user data.