This paper proposes a framework for human action recognition based on procrustes analysis and Fisher vector coding(FVC).Firstly,we applied a pose feature extracted from silhouette image by employing Procrustes analysi...This paper proposes a framework for human action recognition based on procrustes analysis and Fisher vector coding(FVC).Firstly,we applied a pose feature extracted from silhouette image by employing Procrustes analysis and local preserving projection(LPP).Secondly,the extracted feature can preserve the discriminative shape information and local manifold structure of human pose and is invariant to translation,rotation and scaling.Finally,after the pose feature was extracted,a recognition framework based on FVC and multi-class supporting vector machine was employed to classify the human action.Experimental results on benchmarks demonstrate the effectiveness of the proposed method.展开更多
The distance between the two electrode tips can greatly influence the parameters used for record- ing compound nerve action potentials. To investigate the optimal parameters for these recordings in the rat median nerv...The distance between the two electrode tips can greatly influence the parameters used for record- ing compound nerve action potentials. To investigate the optimal parameters for these recordings in the rat median nerve, we dissociated the nerve using different methods and compound nerve action potentials were orthodromically or antidromically recorded with different electrode spac- ings. Compound nerve action potentials could be consistently recorded using a method in which the middle part of the median nerve was intact, with both ends dissociated from the surrounding fascia and a ground wire inserted into the muscle close to the intact part. When the distance be- tween two stimulating electrode tips was increased, the threshold and supramaximal stimulating intensity of compound nerve action potentials were gradually decreased, but the amplitude was not changed significantly. When the distance between two recording electrode tips was increased, the amplitude was gradually increased, but the threshold and supramaximal stimulating intensity exhibited no significant change. Different distances between recording and stimulating sites did not produce significant effects on the aforementioned parameters. A distance of 5 mm between recording and stimulating electrodes and a distance of 10 mm between recording and stimulating sites were found to be optimal for compound nerve action potential recording in the rat median nerve. In addition, the orthodromic compound action potential, with a biphasic waveform that was more stable and displayed less interference （however also required a higher threshold and higher supramaximal stimulus）, was found to be superior to the antidromic compound action potential.展开更多
基金National Natural Science Foundation of China(No.61602148)Natural Science Foundation of Fujian Province,China(No.2016J01040)Xiamen University of Technology High Level Talents Project,China(No.YKJ15018R)
文摘This paper proposes a framework for human action recognition based on procrustes analysis and Fisher vector coding(FVC).Firstly,we applied a pose feature extracted from silhouette image by employing Procrustes analysis and local preserving projection(LPP).Secondly,the extracted feature can preserve the discriminative shape information and local manifold structure of human pose and is invariant to translation,rotation and scaling.Finally,after the pose feature was extracted,a recognition framework based on FVC and multi-class supporting vector machine was employed to classify the human action.Experimental results on benchmarks demonstrate the effectiveness of the proposed method.
基金supported by grants from Hand Function Research Center in Fudan University,Chinathe Project 211 in China,No.211XKZ
文摘The distance between the two electrode tips can greatly influence the parameters used for record- ing compound nerve action potentials. To investigate the optimal parameters for these recordings in the rat median nerve, we dissociated the nerve using different methods and compound nerve action potentials were orthodromically or antidromically recorded with different electrode spac- ings. Compound nerve action potentials could be consistently recorded using a method in which the middle part of the median nerve was intact, with both ends dissociated from the surrounding fascia and a ground wire inserted into the muscle close to the intact part. When the distance be- tween two stimulating electrode tips was increased, the threshold and supramaximal stimulating intensity of compound nerve action potentials were gradually decreased, but the amplitude was not changed significantly. When the distance between two recording electrode tips was increased, the amplitude was gradually increased, but the threshold and supramaximal stimulating intensity exhibited no significant change. Different distances between recording and stimulating sites did not produce significant effects on the aforementioned parameters. A distance of 5 mm between recording and stimulating electrodes and a distance of 10 mm between recording and stimulating sites were found to be optimal for compound nerve action potential recording in the rat median nerve. In addition, the orthodromic compound action potential, with a biphasic waveform that was more stable and displayed less interference （however also required a higher threshold and higher supramaximal stimulus）, was found to be superior to the antidromic compound action potential.
