Seminar

In general, our visual field could be represented by using 10 – 100 MB of information per second. Even if most of neurons are dedicated to such visual processing, it is hard to handle this information in real time. Nevertheless, we can promptly (i.e., computationally efficient) understand a given scene. The most plausible explanation is that the human visual system (HVS) works based on the salient regions in a coarse-to-fine manner. Therefore, it is important for the nervous system to make decisions on which part of the available information is to be selected for further, more detailed processing, and which parts are to be discarded. Computational modeling of this system enables various applications, e.g., image retargeting, object detection and recognition, requiring only limited processing resources. For this reason, a considerable amount of effort has been devoted to detect salient regions, which attract the visual attention indeed, however, previous approaches still suffer from complex backgrounds. In this talk, a novel method for detecting salient regions is presented and its applications to the computer vision field are also introduced.