Home > Published Issues > 2015 > Volume 3, No. 1, June 2015 >

Ultra-Low Bit Rate Facial Coding Hybrid Model Based on Saliency Detection

Wei Zeng, Mingqiang Yang, and Zhenxing Cui
School of Information Science and Engineering, Shandong University, Jinan, China

Abstract—Aiming at getting high quality image sequences in real-time video chat applications at very low bit rate, we propose an improved model-based video coding system in this paper. Instead of detecting human faces with conventional ways, we detect salient regions in the frames using Boolean Map based Saliency (BMS) method to locate the faces. After facial feature extraction by AAM, we transmit the DPCM of the eyes and mouth directly instead of time-consuming facial expression estimation. The new approach we proposed is more suitable for real time and the simulation below indicates the proposed algorithm is effective and feasible.

Index Terms—model-based coding, BMS, saliency detection, facial expression estimation

Cite: Wei Zeng, Mingqiang Yang, and Zhenxing Cui, "Ultra-Low Bit Rate Facial Coding Hybrid Model Based on Saliency Detection," Journal of Image and Graphics, Vol. 3, No. 1, pp. 25-29, June 2015. doi: 10.18178/joig.3.1.25-29