2019/05/16 by Nasrin Sadeghzadehyazdi, Tamal Batabyal, Sadeghzadehyazdi, Nasrin +10
Computer Science · Engineering · Medicine · #Computer Vision and Pattern Recognition (cs.CV) #Diabetic Foot Ulcer Assessment and Management #FOS: Computer and information sciences #FOS: Electrical engineering #Gait Recognition and Analysis #Human Pose and Action Recognition #Image and Video Processing (eess.IV) #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.1905.07058
openalex publication_date 2019/05/16 · openalex created_date 2022/07/29 · openalex updated_date 2026/07/28
Gait recognition using noninvasively acquired data has been attracting an\nincreasing interest in the last decade. Among various modalities of data\nsources, it is experimentally found that the data involving skeletal\nrepresentation are amenable for reliable feature compaction and fast\nprocessing. Model-based gait recognition methods that exploit features from a\nfitted model, like skeleton, are recognized for their view and scale-invariant\nproperties. We propose a model-based gait recognition method, using sequences\nrecorded by a single flash lidar. Existing state-of-the-art model-based\napproaches that exploit features from high quality skeletal data collected by\nKinect and Mocap are limited to controlled laboratory environments. The\nperformance of conventional research efforts is negatively affected by poor\ndata quality. We address the problem of gait recognition under challenging\nscenarios, such as lower quality and noisy imaging process of lidar, that\ndegrades the performance of state-of-the-art skeleton-based systems. We present\nGlidarCo to attain high accuracy on gait recognition under the described\nconditions. A filtering mechanism corrects faulty skeleton joint measurements,\nand robust statistics are integrated to conventional feature moments to encode\nthe dynamic of the motion. As a comparison, length-based and vector-based\nfeatures extracted from the noisy skeletons are investigated for outlier\nremoval. Experimental results illustrate the efficacy of the proposed\nmethodology in improving gait recognition given noisy low resolution lidar\ndata.\n