HPEDHealth and Physical Education
HPEDHealth Promotion and Education Database
HPEDHigh Performance Electric Drive
HPEDHealth Promoting Emergency Departments (Australia)
Copyright 1988-2018 AcronymFinder.com, All rights reserved.
References in periodicals archive ?
We conducted extensive experiments on CK+ for the given 7-expression classes to show the comparative efficacy of HPED over existing descriptors and other state-of-the-art methods.
We conducted a person-independent expression recognition for this noise-corrupted dataset and compared the results of HPED against other descriptors, including LBP [1], LDN [4], LDP [7], LPTP [8], PTP [9], HOG [12], and LDTP [13].
Performance under positional variation (misalignment): We evaluated the performance of HPED after distorting the alignment of the frontal face to test its robustness under a subtle registration error when detecting the frontal face.
Performance under low-resolution: In addition, we investigated the performance of HPED descriptors in different resolution images to test its comparative efficacy in such images, especially in the low-resolution images.
In particular, we compared the performance of HPED against different appearance-based methods and deep learning based methods.
In addition, we provide the confusion matrix of HPED for 7-class CK+ images in Fig.
Expression recognition results for all the FACES images are presented in the first column of Table 5, which shows that the HPED descriptor achieves better performance against the existing local descriptors.
Nevertheless, the proposed HPED shows superior performance against other descriptors, which points to the efficacy of the proposed feature extraction scheme of HPED when applied to such a challenging dataset with large variations.
Hence, such higher performance can also be interpreted as the efficacy of HPED for race variations.
To this end, one may ask why only two directions were used, that is, why the primary and secondary directions are generated in the HPED descriptor.