Modeling Subbands of a Wavelet Based Scalable Video Codec

In the RESUME project we explore the use of reconfigurable hardware for the design of portable multimedia systems by developing a hardware-friendly scalable wavelet-based video codec. Our scalable video codec provides the ability to decoded the video stream with reduced frame rate, resolution or image quality directly from the original encoded video stream. This is important for portable devices that have different Quality of Service (QoS) requirements and power restrictions. Our video codec makes use of arithmetic coding to achieve compression by exploiting statistical redundancy in video streams. This technique employs modeling for building its statistical information. In this paper we explore the use of wavelet subband models. We demonstrate that even if the available memory is extremely limited we come very close to the optimal compression. Key-Words: scalable video, wavelet, arithmetic coding, context modeling