Video compression using matching pursuits

Abstract

Abstract-The use of matching pursuit (MP) to code video using overcomplete Gabor basis functions has recently been introduced. In this paper, we propose new functionalities such as SNR scalability and arbitrary shape coding for video coding based on matching pursuit. We improve the performance of the baseline algorithm presented earlier by proposing a new search and a new position coding technique. The resulting algorithm is compared to the earlier one and to DCT-based coding.

Efficient Scalable Video Coding Based on Matching Pursuits

“…In our study, we focus on developing SNR scalability algorithms at low bit rates in a hybrid motion-compensation video coding system in which the frame residuals are encoded using matching pursuits. Certain matching pursuit SNR-scalable schemes have been proposed in (Al-Shaykh et al, 1999;Vleeschouwer & Macq, 2000). An FGS produces a continuous bit-stream with increasing PSNR for a wide range of bit rates.…”

“…Thus, the matching pursuit update algorithm cannot be used in video encoding. Instead, the proposed approach in (Neff & Zakhor, 1997;Al-Shaykh et al, 1999) divides a residual into blocks and, at each iteration, the matching pursuit algorithm is applied to the block with the highest energy. This approach is both simple and efficient, and has been implemented in many MP-based video codecs.…”

“…Even so, current approaches based on hybrid motion compensation and residual encoding schemes using the DCT are still inefficient, especially at low bit-rates. Scalable video coding schemes based on matching pursuit have been proposed as a means of achieving scalability (Al-Shaykh et al, 1999;Vleeschouwer & Macq, 2000). In (Al-Shaykh et al, 1999), scalability is supported by successively en-coding groups of atoms; thus, it is constricted by the number of atoms determined by the encoder.…”

“…Scalable video coding schemes based on matching pursuit have been proposed as a means of achieving scalability (Al-Shaykh et al, 1999;Vleeschouwer & Macq, 2000). In (Al-Shaykh et al, 1999), scalability is supported by successively en-coding groups of atoms; thus, it is constricted by the number of atoms determined by the encoder. Generally, providing video coding with scalability degrades the coding performance.…”

“…Because the matching pursuit algorithm is a data-dependent frame-based representation, a matching pursuit video coding technique cannot be directly translated from conventional transform-based approaches. We must therefore develop a new matching pursuit video coding technique that can deal with quantization noise in the matching pursuit algorithm (Neff & Zakhor, Sept. 2000;Vleeschouwer & Zakhor, 2002), multiple description coding for reliable transmission (Tang & Zakhor, 2002), scalable bitstream generation (Al-Shaykh et al, 1999;Vleeschouwer & Macq, 2000;Rose & Regunathan, 2001), and dictionary learning and adaptation (Engan et al, 2000).…”