Multi-scale based motion estimation for high quality video coding

Block-based motion estimation (ME) has been widely used in various video coding standards due to its effectiveness removing temporal redundancy. It plays a key role in the transmission and storage of video signals at reduced bit rates. In this paper, a novel motion estimation algorithm is proposed to enhance video quality by reflecting enlarging and reducing movements such as zoom in/out scene. The experimental results show that the proposed algorithm achieves average quality enhancement of 1.19 dB (up to 2.34 db) in PSNR gain with acceptable increment of ME time.