Digital image stabilization with sub-image phase correlation based global motion estimation

Ertürk, Şükrü Mehmet

doi:10.1109/tce.2003.1261235

Cited by 160 publications

(89 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Motion estimation methods can be classified into several types. The three common types of motion estimation are block matching [5], [6] phase correlation [7]- [9], and gradient field correlation [10], [11]. Each of the methods has its own advantages and disadvantages.…”

Section: Video Compression and Motion Estimationmentioning

confidence: 99%

“…Block based methods have been widely used in many applications in motion estimation and can be implemented without much difficulty as compared to other methods. Gradient field [12] and phase correlation method [13] are promising methods but they require a lot of effort to calculate the gradient field and or phase information of signal respectively.…”

Section: Video Compression and Motion Estimationmentioning

confidence: 99%

See 1 more Smart Citation

A Motion Estimation Algorithm Using DTCWT and ARPS

Oktiawati

Yap

2012

itbj.ict.

View full text Add to dashboard Cite

Abstract. In this paper, a hybrid motion estimation algorithm utilizing the Dual Tree Complex Wavelet Transform (DTCWT) and the Adaptive Rood Pattern Search (ARPS) block is presented. The proposed algorithm first transforms each video sequence with DTCWT. The frame n of the video sequence is used as a reference input and the frame n+2 is used to find the motion vector. Next, the ARPS block search algorithm is carried out and followed by an inverse DTCWT. The motion compensation is then carried motion vector. The results show that PSNR can be improved for mobile device without depriving its quality. The proposed algorithm also takes less memory usage compared to the DCT is a hybrid wavelet-based motion estimation algorithm for mobile devices. Other contribution is the visual quality scoring system as used in section 6. Keywords: ARPS; DTCWT; mobile device; PSNR; video compression. IntroductionIn recent years, the transmission of multimedia content using mobile devices has become very popular. Teleconference, video streaming and digital television are just some examples of its applications. The demand for multimedia content via mobile devices is getting higher. Th result in heavier demands on the radio spectrum and hence, air bandwidth will remain a bottleneck. In this paper, a hybrid motion estimation algorithm utilizing the Dual Tree Complex Wavelet Transform (DTCWT) and the Adaptive Rood Pattern Search (ARPS) block is presented. The proposed algorithm first transforms each e with DTCWT. The frame n of the video sequence is used as a reference input and the frame n+2 is used to find the motion vector. Next, the ARPS block search algorithm is carried out and followed by an inverse DTCWT. The motion compensation is then carried out on each inversed frame n and motion vector. The results show that PSNR can be improved for mobile device without depriving its quality. The proposed algorithm also takes less memory usage compared to the DCT-based algorithm. The main contribution of this work based motion estimation algorithm for mobile devices. Other contribution is the visual quality scoring system as used in section 6. ARPS; DTCWT; mobile device; PSNR; video compression.e transmission of multimedia content using mobile devices has become very popular. Teleconference, video streaming and digital television are just some examples of its applications. The demand for multimedia content via mobile devices is getting higher. This will invariably result in heavier demands on the radio spectrum and hence, air bandwidth will Uncompressed video data usually need considerable data storage capacity and data transmission bandwidth. Furthermore, uncompressed video data can influence transmission time needed. Large space requirement, high computational cost, and long transmission time are only some problems in transmitting video or multimedia content. For example, transmitting a 5 minute PAL video via a modem at 56 kilobits per second would take about 91.2 hours e with DTCWT. The frame n of the video sequence is used as a reference...

show abstract

Section: Video Compression and Motion Estimationmentioning

confidence: 99%

Section: Video Compression and Motion Estimationmentioning

confidence: 99%

A Motion Estimation Algorithm Using DTCWT and ARPS

Oktiawati

Yap

2012

itbj.ict.

View full text Add to dashboard Cite

show abstract

“…Ertürk et al [2] stated that image stabilization may remove all of the detected global motion, but is not desired and would remove intentional movements as well. This may be applicable for image stabilization in consumer cameras, but the vibration mentioned in this paper has to be removed completely.…”

Section: Introductionmentioning

confidence: 99%

Vibration Compensation for Fisheye Lenses in UAV Applications

Gurtner

Walker

Boles

2007

9th Biennial Conference of the Australian Pattern Recognition Society on Digital Image Computing Techniques and Applications (D

View full text Add to dashboard Cite

show abstract

“…Другой -для каждой текущей оценки вектора смещения пере-брать все соседние целочисленные смещения, например, в окрестности 3×3, и выбрать лучший вариант следующей оценкой [6]. Подход на основе корреляции фаз описан в [3,4].…”

unclassified