The arbitrarily shaped transform of segmented motion field for a pseudo object-oriented very low bit-rate video coding system

The arbitrarily shaped transform (AST) of the segmented motion field generated by modified optical flow algorithm (MOFA) is suggested in this paper. Because the transform kernels are dependent on the shape of the coded region, information about the contours of the objects must be transmitted. To achieve data compression, a coefficient selection function (CSF) is defined for AST coefficients reduction. Since shape information data intensive, a rectangle approximation method is also developed. According to the simulation results, the proposed coding method is efficient and its performance is much better than the traditional block matching algorithm in terms of PSNR and bit-rate. In addition, the proposed algorithms are integrated together to build a very low bit-rate video coding system. From some primary simulations of the system, the proposed method can guarantee 34 dB PSNR for most cases at around 10 kbps; furthermore, the segmentation of objects has potential for object scalability