This paper proposes using computer vision approaches based on local feature representation, feature learning, and classification for sex prediction from human cranial data obtained from CT scans.
Using an algorithm based on local over-representation and comparative genomics, we identified putatively functional transcription factor binding sites (TFBS) in TATA-dependent proximal-promoters.
We therefore summarize the contributions of the paper as follows: This paper proposes using computer vision approaches based on local feature representation, feature learning, and classification for sex prediction from human cranial data obtained from CT scans.
The boundary grid is generated either by a one-dimensional version of the advancing-front method, or by requiring a set of criteria based on local line representation with circular arcs.
Although most quantization-based global representations are based on local image descriptors (such as SIFT [2] and SURF [12]), a few important spatial relevant information of these local features are generated by the local feature detectors [13], such as coordinates, scale, orientation, or saliency of local feature points apart from the local feature descriptors.
Therefore, we directly concatenate the image representations based on local distance vector and local features to achieve superior performance.
Since the rub fault is highly nonlinear and transient, time-frequency representation (TFR) based on local mean decomposition (LMD) is extensively used for the study.
The selection of the monitors was based on local criteria, mainly on the completeness of measurements and representation of population exposure.
Compared to learning based on local generalizations, the number of patterns that can be obtained using a distributed representation scales quickly with the number of learnt factors.
In most CBIR systems, global image representations (such as BoW [3], VLAD [8], and FV [10]) are based on local image descriptors, whereas the output of the feature detector information is ignored.
