Video coding by applying the extension of EZW to 3D and the lifting scheme

• Author(s): Salvado, José ; Sousa, Leonel
• Date: 2002
• Persistent ID: http://hdl.handle.net/10400.11/107
• Origin: Repositório Científico do Instituto Politécnico de Castelo Branco
• Subject(s): Lifting scheme; Video coding; Wavelet transform
• Document Type: Conference Object
• Language: English

Description

This paper proposes a video codec based on the discrete wavelet transform (DWT). The video sequences are first splitted in groups of frames(GOF). Each GOF is then decomposed in the time domain, and in the spatial domain (for each frame) by applying the DWT. We use the lifting scheme to compute the DWT coefficients. The coefficients of the 3D structure resulting from both DWT (in time and space domain) are than coded througth the extension of EZW coding to 3D and the bitstream is finnaly passed through a Huffman coder, to achieve better compression ratios. The structure of the video codec and some experimental results are presented. Experimental results evaluate the quality of the codec images and compression ratio achieved with the proposed codec.