In trecvid 2010, our team takes part in 4 event detection competition including embracing, pointing, object put and cell to ear. We build four systems to recognize these events separately. For embracing, we use a probability accumulated method. For pointing, we use motion energy image and rules to recognize this action. For object put, we use probability rule. And for cell to ear, we use orient...

Optical flow can be used to segment a moving object from its background provided the velocity of the object is distinguishable from that of the background, and has expected characteristics. Existing optical flow techniques often detect flow (and thus the object) in the background. To overcome this, we propose a new optical flow technique, which only determines optical flow in regions of motion....

Visual servo can be achieved by extracting motion information from the optical flow, defined as the apparent motion of brightness patterns of an image. Control performance using this technique fully depends on accurate estimation of the flow velocities. It appears that most correct values of the optical flow usually locate at the regions with image features. In this paper, feature-based optical...

Please cite this article in press as: L. Li, J.C.K. Ch j.displa.2012.10.005 We have shown that people steer toward a target by aligning their heading with the target when target egocentric direction is not available for steering [24]. Here we examined what visual strategies people use to steer toward a target when target egocentric direction is available for steering. The display simulated a pa...

The measurement of optical ow is a fundamental problem in Computer Vision. Many techniques have been presented in the literature and many more continue to appear. How would one select an appropriate method for a given task? We overview the various classes of optical ow methods and describe examples of each, emphasizing both the advantages and drawbacks of each class of methods. We conclude the ...

Dense trajectories have been shown as a very promising method in the human action recognition field. In this paper, we propose a new kind of video descriptor, generated from the relationship between the trajectory’s optical flow with the gradient field in its neighborhood. Orientation tensors are used to accumulate relevant information over the video, representing the tendency of direction in t...