Motion vector prediction for improving one bit transform based motion estimation

Author(s):  
Colin Doutre ◽  
Panos Nasiopoulos
2008 ◽  
Vol 5 (21) ◽  
pp. 889-894
Author(s):  
Jinha Choi ◽  
Wonjae Lee ◽  
Yunho Jung ◽  
Jaeseok Kim

2014 ◽  
Vol 556-562 ◽  
pp. 4365-4371
Author(s):  
Ming Hui Yang ◽  
Xiao Dong Xie

Motion Vector Prediction (MVP) plays an important role in improving coding efficiency in HEVC, H.264/AVC and AVS video coding standard. MVP is implemented by exploiting redundancy of adjacent-block optimal coding information under the constraint that MVP must be performed in a serial way. The constraint prevents parallel processing and MB pipeline based on LevelC+. In multi-stage pipeline, to some extent, adjacent-block best mode-decision information can hardly be obtained. In this paper, we propose a new hardware-oriented method to improve the coding performance at a cost of few hardware resources. When adjacent block is not available, spatial motion vector prediction (SMVP) for integer motion estimation (IME) and fraction motion estimation (FME) will take the IME best mode information and FME best mode information of left block as best information to derive PMV (Predicted Motion Vector) for current macro-block or block. Experimental results shows that the method we propose can achieve a better performance than the existing methods by 0.1db for the cases with intense movement and a non-degrading performance for flat cases.


2011 ◽  
Vol 145 ◽  
pp. 277-281
Author(s):  
Vaci Istanda ◽  
Tsong Yi Chen ◽  
Wan Chun Lee ◽  
Yuan Chen Liu ◽  
Wen Yen Chen

As the development of network learning, video compression is important for both data transmission and storage, especially in a digit channel. In this paper, we present the return prediction search (RPS) algorithm for block motion estimation. The proposed algorithm exploits the temporal correlation and characteristic of returning origin to obtain one or two predictive motion vector and selects one motion vector, which presents better result, to be the initial search center. In addition, we utilize the center-biased block matching algorithms to refine the final motion vector. Moreover, we used adaptive threshold technique to reduce the computational complexity in motion estimation. Experimental results show that RPS algorithm combined with 4SS, BBGDS, and UCBDS effectively improves the performance in terms of mean-square error measure with less average searching points. On the other hand, accelerated RPS (ARPS) algorithm takes only 38% of the searching computations than 3SS algorithm, and the reconstruction image quality of the ARPS algorithm is superior to 3SS algorithm about 0.30dB in average overall test sequences. In addition, we create an asynchronous learning environment which provides students and instructors flexibility in learning and teaching activities. The purpose of this web site is to teach and display our researchable results. Therefore, we believe this web site is one of the keys to help the modern student achieve mastery of complex Motion Estimation.


Sign in / Sign up

Export Citation Format

Share Document