Gait-Based Gender Classification Using Persistent Homology

Abstract

Abstract. In this paper, a topological approach for gait-based gender recognition is presented. First, a stack of human silhouettes, extracted by background subtraction and thresholding, were glued through their gravity centers, forming a 3D digital image I. Second, different filters (i.e. particular orders of the simplices) are applied on ∂K(I) (a simplicial complex obtained from I) which capture relations among the parts of the human body when walking. Finally, a topological signature is extracted from the persistence diagram according to each filter. The measure cosine is used to give a similarity value between topological signatures. The novelty of the paper is a notion of robustness of the provided method (which is also valid for gait recognition). Three experiments are performed using all human-camera view angles provided in CASIA-B database 1 . The first one evaluates the named topological signature obtaining 98.3% (lateral view) of correct classification rates, for gender identification. The second one shows results for different human-camera distances according to training and test (i.e. training with a human-camera distance and test with a different one). The third one shows that upper body is more discriminative than lower body.

Topological Data Analysis for Image Tampering Detection

“…These three important properties of topology have been discussed in detail in [7,8]. Recent application scope of topological data/image analysis includes, but is not limited to, gait recognition [9,10,22], brain artery [20], hurricane and galaxies analysis [11], dimensionality reduction schemes evaluation [12], classification of hepatic lesions [21], shape classification using LBP and persistent [13] and many more.…”

Gait-Based Carried Object Detection Using Persistent Homology

“…This new approach allows to decrease the number of maximal cells to the half respect to the usually applied simplicial complexes [9,10], achieving a much better efficiency without sacrificing the efficacy.…”

“…The 3D binary digital picture I = (Z 3 , B) (where B ⊂ Z 3 is the foreground), is built by stacking silhouettes of a subsequence of representation, aligned by their gravity centers (gc) [9,10]. See Fig.…”

“…We build models of naturally-walking and object-carrying using persistent homology groups. These features have been previously used for gait identification and gender classification [9,10]. Our methodology is evaluated using the CASIA-B gait database 1 , which consists of 738 video sequences of persons walking naturally, without carrying object and 246 video sequences of persons walking with different kinds of bags.…”