Combining Key Frame Based Motion Design with Controlled Movement Execution

In this paper, we study the challenge of image-to-video retrieval, which uses the query image to search relevant frames from a large collection of videos. A novel framework based on convolutional neural networks (CNNs) is proposed to perform large-scale video retrieval with low storage cost and high search efficiency. Our framework consists of the key-frame extraction algorithm and the feature aggregation strategy. Specifically, the key-frame extraction algorithm takes advantage of the clustering idea so that redundant information is removed in video data and storage cost is greatly reduced. The feature aggregation strategy adopts average pooling to encode deep local convolutional features followed by coarse-to-fine retrieval, which allows rapid retrieval in the large-scale video database. The results from extensive experiments on two publicly available datasets demonstrate that the proposed method achieves superior efficiency as well as accuracy over other state-of-the-art visual search methods.

Download Full-text

Gait recognition based on key frame and elliptical model

The 2010 IEEE International Conference on Information and Automation ◽

10.1109/icinfa.2010.5512194 ◽

2010 ◽

Author(s):

Junhong Xu ◽

Wang Cong ◽

Jin Li ◽

Lei Wang ◽

Lijie Li ◽

...

Keyword(s):

Gait Recognition ◽

Key Frame ◽

Elliptical Model

Download Full-text

Facilitation of cutaneous inputs during the planning phase of gait initiation

Journal of Neurophysiology ◽

10.1152/jn.00668.2014 ◽

2015 ◽

Vol 114 (1) ◽

pp. 301-308 ◽

Cited By ~ 11

Author(s):

Laurence Mouchnino ◽

Aurélie Fontan ◽

Christophe Tandonnet ◽

Joy Perrier ◽

Anahid H. Saradjian ◽

...

Keyword(s):

Evoked Potential ◽

Gait Initiation ◽

Motor Execution ◽

Movement Execution ◽

Movement Initiation ◽

Somatosensory Stimulation ◽

Planning Phase ◽

Step Initiation ◽

Afferent Information ◽

Afferent Inputs

It has been shown that during the planning of a voluntary movement the transmission of cutaneous afferent inputs to the somatosensory cortex is attenuated shortly before the motor output as well as during movement execution. However, it is not known whether the sensory suppression observed during the planning phase (i.e., before any movement execution) is a systemic phenomenon or whether it is dependent on movement context. For example, movements such as step initiation are controlled based on information received from cutaneous receptors in the feet. Because afferent information emerging from these receptors is critical for movement initiation, we hypothesized that suppression of these inputs may not occur during the planning phase prior to gait initiation. To examine this hypothesis we measured the cortical response to somatosensory stimulation during the planning phase of step initiation and during movement execution. Sensitivity to cutaneous stimulation was assessed by measuring the amplitude of the cortical somatosensory-evoked potential (SEP, over the Cz electrode) following electrical stimulations of the plantar sole of one foot. Two stimulations were provided during the planning phase of a step movement and two stimulations during movement execution. It was found that the P50-N80 SEP was facilitated in the early planning phase (−700 ms before motor execution) compared with when participants remained still (control standing task). This mechanism might contribute to an enhanced perception of cutaneous input leading to a more accurate setting of the forces to be exerted onto the ground to shift the body's weight toward the supporting side prior to foot-off.

Download Full-text