Global Motion Estimation and Its Applications

Qian, Xueming

doi:10.5772/34166

Cited by 3 publications

(4 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We continue iterations until the element-wise L 2 matrix norm ||Ĥ k´I3 || 2 gets smaller than a convergence threshold , or a maximal number of iterations K max is reached. To estimate large-scale global motion efficiently, we apply Algorithm 1 in a coarse-to-fine manner through a pyramid representation [40]. If K max is large enough, a sufficient condition for convergence is that the error between the estimated and the actual homography transformation, ||Ĥ´H|| 2 , is reduced at each iteration k, which implies that the MVs must be estimated more accurately as k increases.…”

Section: B Global Motion Estimationmentioning

confidence: 99%

“…To evaluate GME performance in the presence of outlier motions, we adopt a popular validation scheme [40] and use test sequences that contain mostly camera motion in addition to outlier local motions. The goal is to compensate the global motion between pairs of consecutive images by warping the second (target) frame in each pair onto the first (template) frame.…”

Section: Global Motion Estimation Experimentsmentioning

confidence: 99%

“…LK-based approaches tolerate outlier motions, as the contribution of each local image patch to the optimised energy function is independent from the rest of the image. However, if outlier local motions occur in a large portion of the image, their contribution to the energy function may become significant and undermine the accuracy of GME [40]. Numerous extensions of LK have been proposed, which differ in the energy function that is optimised, the optimisation algorithm that is employed or the domain where the optimisation is performed [34], [36], [41], [35], [42], [43].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Biologically Inspired Motion Encoding for Robust Global Motion Estimation

Sarıyanidi

Güneş

Cavallaro

2017

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

The growing use of cameras embedded in autonomous robotic platforms and worn by people is increasing the importance of accurate global motion estimation (GME). However, existing GME methods may degrade considerably under illumination variations. In this paper, we address this problem by proposing a biologically-inspired GME method that achieves high estimation accuracy in the presence of illumination variations. We mimic the early layers of the human visual cortex with the spatio-temporal Gabor motion energy by adopting the pioneering model of Adelson and Bergen and we provide the closed-form expressions that enable the study and adaptation of this model to different application needs. Moreover, we propose a normalisation scheme for motion energy to tackle temporal illumination variations. Finally, we provide an overall GME scheme which, to the best of our knowledge, achieves the highest accuracy on the Pose, Illumination, and Expression (PIE) database.

show abstract

Section: B Global Motion Estimationmentioning

confidence: 99%

Section: Global Motion Estimation Experimentsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Biologically Inspired Motion Encoding for Robust Global Motion Estimation

Sarıyanidi

Güneş

Cavallaro

2017

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

show abstract

“…According to [16], a zoom is then detected if m 0 and m 4 are equals. We set then a threshold on their difference.…”

Section: Zoom Detectionmentioning

confidence: 99%

Singlets: Multi-resolution Motion Singularities for Soccer Video Abstraction

Blanc

Lingrand

Precioso

2017

2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)

View full text Add to dashboard Cite

The burst of video production appeals for new browsing frameworks. Chiefly in sports, TV companies have years of recorded match archives to exploit and sports fans are looking for replay, summary or collection of events.In this work, we design a new multi-resolution motion feature for video abstraction. This descriptor is based on optical flow singularities tracked along the video. We use these singlets in order to detect zooms, slow-motions and salient moments in soccer games and finally to produce an automatic summarization of a game.We produce a database for soccer video summarization composed of 4 soccer matches from HDTV games for the FIFA world cup 2014 annotated with goals, fouls, corners and salient moments to make a summary. We correctly detect 88.2% of saliant moments using this database. To highlight the generalization of our approach, we test our system on the final game of the handball world championship 2015 without any retraining, refining or adaptation.

show abstract

Sensor assisted Motion Estimation

Areekath

Palavalasa

2013

2013 Students Conference on Engineering and Systems (SCES)

View full text Add to dashboard Cite

Linear displacement estimation of a mobile device is useful in various real time applications, such as Global Motion Estimation (GME) in video encoding, video surveillance etc. Present state of the art techniques perform the global motion estimation using techniques predominantly based on image processing algorithms. These require complex computations and more power consumption.In this paper a novel approach, Sensor assisted Motion Estimation (SaME), to estimate the linear displacement of a mobile device using inbuilt sensors, is proposed. SaME uses inbuilt 3-axis accelerometer to determine the linear displacement along X, Y and Z axes. SaME algorithm is a combination of basic image processing techniques, sensor data and Artificial Intelligence. The Artificial Neural Network (ANN) is used to build the artificial intelligence. ANN is trained to map the sensor data from inbuilt accelerometer to the corresponding displacement data acquired via simple image processing techniques. The resultant model is used to estimate the linear displacement of the mobile device using the sensor data as input. From the results it can be concluded that the inbuilt sensors can be used to arrive at reliable estimates of linear displacement via the proposed method.

show abstract

Global Motion Estimation and Its Applications

Cited by 3 publications

References 23 publications

Biologically Inspired Motion Encoding for Robust Global Motion Estimation

Biologically Inspired Motion Encoding for Robust Global Motion Estimation

Singlets: Multi-resolution Motion Singularities for Soccer Video Abstraction

Sensor assisted Motion Estimation

Contact Info

Product

Resources

About