Source model for transform video coder and its application. I. Fundamental theory

Abstract

A source model describing the relationship between bits, distortion, and quantization step sizes of a large class of block-transform video coders is proposed. This model is initially derived from the rate-distortion theory and then modified to match the practical coders and real image data. The realistic constraints such as quantizer dead-zone and threshold coefficient selection are included in our formulation. The most attractive feature of this model is its simplicity in its final form. It enables us to predict the bits needed to encode a picture at a given distortion or to predict the quantization step size at a given bit rate. There are two aspects of our contribution: one, we extend the existing results of rate-distortion theory to the practical video coders, and two, the nonideal factors in real signals and systems are identified, and their mathematical expressions are derived from empirical data. One application of this model, as shown in the second part of this paper, is the buffer/quantizer control on a CCITT P 2 64 k coder with the advantage that the picture quality is nearly constant over the entire picture sequence.

An Algorithm for Learning Orthonormal Matrix Codebooks for Adaptive Transform Coding

“…In coding of image, video, and other realworld signals, transform coefficients are typically modeled by unimodal pdfs such as Gaussian or Laplace pdfs [9], [33]- [35]. In the following, we obtain closed-form expressions for…”

Rate control for H.264/AVC using enhanced coding complexity measure

“…Extensive simulation results show that the DCT coefficients distribution model in [10] is more generic and suitable for low distortion video compression applications, whose R-D function is as follows: show that this point is at 2 / ε ρe . Therefore, the R-D function (1) can be expanded into a Taylor series at 2 / ε ρe as follows:…”

Convex Optimization-Based Bit Allocation for Video Coding

“…Consequently, the practical effectiveness of the analytic entropy expressions in traditional R-D analysis is somewhat limited. Therefore, a variety of empirical R-D models are developed [8][9][10][11][12][20][21][22] in practical DCTbased image/video coding. Many models simply treat the distortion as a linear function of the quantizer, or even as the quantizer value itself.…”

“…Traditional R-D analysis works [21,23,24] try to seek the relationship between the bit rate, which is usually approximated by the zero-order entropy of the quantized DCT coefficients, and the resulting quantization error (distortion) based on some typical DCT coefficient distributions [17][18][19]. However, considerable correlation among the quantized coefficients still exists, which can be further reduced before entropy coding by zig-zag scan and differential encoding of DC coefficients [21]. This usually results in a large discrepancy between the zero-order entropy and the actual bits used.…”