Conference Paper

A core for ambient and mobile intelligent imaging applications

Inst. of Microelectron. Syst., Hannover Univ., Germany
DOI: 10.1109/ICME.2003.1221538 Conference: Multimedia and Expo, 2003. ICME '03. Proceedings. 2003 International Conference on, Volume: 2
Source: IEEE Xplore

ABSTRACT This paper describes the work in progress of the European IST-2001-34410 CAMELLIA project, which focuses on a platform-based development of a smart imaging core to be embedded in smart cameras. Therefore the work within the CAMELLIA project comprises the specification and implementation of smart imaging applications including the development of required new algorithms and hardware. Based on an existing video encoding architecture for MPEG-4 Simple Profile, the aim is to design a smart imaging core, which is suitable for automotive and mobile communication applications. Thereby the encoding architecture is to be extended with processing units for low- and mid-level smart imaging functions. To indicate the applicability of this platform-based development, a first approach for a motion estimation based background detection using a hardware motion estimation unit is illustrated in this paper. Furthermore, first results employing the background detection to detect moving objects in video sequences using a functional simulator of a video encoding architecture are presented.

  • Source
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper documents the development of a camera that combines real-time video encoding with the tracking of a moving non-rigid object. The object tracker reuses motion vectors that are generated for video encoding by grouping similar vectors together. These motion vectors are produced using block matching motion estimation. A rectangular representation of the moving object is then formed using these block clusters. This system was first implemented as a PC-based prototype as a proof of concept. It was then ported onto an Altera Nios II soft- core processor where some hardware-based support was used to be able to meet application timing requirements. Practical tests have shown that it is feasible to extend an embedded real-time video encoder to track non-rigid objects using the prototype developed in this project.
  • [Show abstract] [Hide abstract]
    ABSTRACT: We present an improvement to the object detection scheme of Viola and Jones (5), using the example of face detec- tion. We present a new kind of visual features, which is based on sampling individual pixels within the examined sub-window, instead of comparing the sums of pixel val- ues in rectangular regions. Our features are faster and do not demand the preparation of an integral image, nor variance-normalization of each sub-window. The result is an improved system which detects faces in quarter-PAL im- ages in only 9 millisecond per image on a 2.4GHz pentium computer.