The capability to recognize objects in online mode is an important aspect of intelligence in multimedia systems. Online object recognition provided by camera device enables video indexing to be done at camera site, which improves greatly architectural possibilities concerning material recording, search and retrieval. Classification of object at camera site enables automatic reactions concerning e.g. recording resolution or compression parameters adjustments. In multimedia systems object recognition capable camera has great potential of improving human –computer interface communication, including human-like automated decision making, i.e., automatic navigation and control tools. However, applying online object recognition requires not only efficient object recognition to be developed but they also demand near-real-time processing speed and optimization to limited resources of computational chips. The goal of this article is to review the challenge of online object recognition on camera device, review available object recognition methods, and address their applicability in the context. Moreover, we review the issues related to using image descriptors, object definitions and object recognition in given context of online processing applied on video camera.
CITATION STYLE
Karimaa, A. (2014). A survey of hardware accelerated methods for intelligent object recognition on camera. In Advances in Intelligent Systems and Computing (Vol. 240, pp. 523–530). Springer Verlag. https://doi.org/10.1007/978-3-319-01857-7_51
Mendeley helps you to discover research relevant for your work.