This volume constitutes the refereed proceedings of the Second International Conference on Multimedia and Signal Processing, CMSP 2012, held in Shanghai, China, in December 2012.<br>The 79 full papers included in the volume were selected from 328 submissions from 10 different countries and regions. The papers are organized in topical sections on computer and machine vision, feature extraction, image enhancement and noise filtering, image retrieval, image segmentation, imaging techniques &amp; 3D imaging, pattern recognition, multimedia systems, architecture, and applications, visualization, signal modeling, identification &amp; prediction, speech &amp; language processing, time-frequency signal analysis.