Temporally adaptive motion estimation using temporal subband analysis

In this paper, a new scheme that measures the significance of temporal activities of moving objects in image sequence is proposed. The measured temporal activity distribution is then used for motion estimation in MPEG like video coder. The proposed `adaptive motion estimation´ scheme consists of five function blocks: temporal subband analysis (TSBA), extraction of temporal activity distribution (TAD), scene change detection (SCD), picture type replacement (PTR), and temporally adapted block matching algorithm (TABMA). All five function blocks are developed primarily based on the temporal subband analysis. According to the significance of TAD, TABMA adjusts its nonuniform motion vector searching area. Also, newly proposed SCD and PTR together can prevent poor motion prediction for a video sequence containing abruptly changed scenes in it. Computer simulation results show that the proposed motion estimation scheme is superior to the one of typical MPEG like video coders: it requires less computations, and it shows better performances