| dc.description.abstract | Foreground segmentation for video frames has been an important role in video surveillance, pattern recognition, video indexing, and video coding. Due to the large amount of video data, videos have to be compressed before storage and transmission. Foreground segmentation based on compression information saves the processing to the original frame, therefore, is an algorithm suitable for real-time applications.
In recent years, video compression standards had been promoted rapidly. In the H.264/AVC video coding standard, in addition to motion vectors, there are also seven-mode block partitions which can provide extra information for segmentation. In former algorithms for moving object segmentation for video acquired by moving cameras, they first approximated the relative global motion model using all motion vectors, than marked the blocks with motion vectors differed from the global motion by an amount as foreground blocks. During the procedure described above, according to the different block partition modes, we can choose the preferable MVs for estimating the global motion, moreover, improve the accuracy of the judgments on blocks as being foreground or background. Finally, we refine the results with spatial and temporal filters of our design and segment foreground with proposed adaptive threshold.
With the use of motion vectors, we can process the video data from moving cameras. It makes this algorithm more practical than many object segmentation methods using spatial domain information. It also reduces the computational costs and can be used for real-time systems. | en_US |