Multiple Description Scalar Quantization Based Image Coding using Optimal Filtering

This paper presents an image coding scheme based on multiple description scalar quantization which is one of the first practical multiple description image coding methods. The proposed approach aims to estimate lost descriptions at the decoder in least squares sense using optimal filter coefficients obtained at the encoder. The obtained optimal filter coefficients are multiplexed into each description at the encoder and used to reconstruct lost description at the decoder to improve overall coding efficiency. Experimental results show that the proposed approach improves performance of the conventional multiple description scalar quantization approach.