Relative effects of posture and activity on human height estimation from surveillance footage.

Height estimations based on security camera footage are often requested by law enforcement authorities. While valid and reliable techniques have been established to determine vertical distances from video frames, there is a discrepancy between a person's true static height and their height as measured when assuming different postures or when in motion (e.g., walking). The aim of the research presented in this report was to accurately record the height of subjects as they performed a variety of activities typically observed in security camera footage and compare results to height recorded using a standard height measuring device. Forty-six able bodied adults participated in this study and were recorded using a 3D motion analysis system while performing eight different tasks. Height measurements captured using the 3D motion analysis system were compared to static height measurements in order to determine relative differences. It is anticipated that results presented in this report can be used by forensic image analysis experts as a basis for correcting height estimations of people captured on surveillance footage.

[1]  D Casey Kerrigan,et al.  Gender differences in pelvic motions and center of mass displacement during walking: stereotypes quantified. , 2002, Journal of women's health & gender-based medicine.

[2]  Ivo Alberink,et al.  Comparison of the Performance of Two Methods for Height Estimation , 2010, Journal of forensic sciences.

[3]  Barbara S Smith,et al.  Influence of carrying a backpack on pelvic tilt, rotation, and obliquity in female college students. , 2006, Gait & posture.

[4]  Ye-Peng Guan Unsupervised human height estimation from a single image , 2009 .

[5]  Jack C Y Cheng,et al.  The effect of backpack load on the gait of normal adolescent girls , 2005, Ergonomics.

[6]  R L Waters,et al.  Electrical activity of muscles of the trunk during walking. , 1972, Journal of anatomy.

[7]  Larry S. Davis,et al.  Person identification using automatic height and stride estimation , 2002, Object recognition supported by user interaction for service robots.

[8]  R. Baker Gait analysis methods in rehabilitation , 2006, Journal of NeuroEngineering and Rehabilitation.

[9]  J. Hoffman,et al.  American Journal of Physical Medicine and Rehabilitation , 2007 .

[10]  J. Bloomberg,et al.  Identifying head-trunk and lower limb contributions to gaze stabilization during locomotion. , 2002, Journal of vestibular research : equilibrium & orientation.

[11]  Viviana Rota,et al.  Gait Analysis on Split-Belt Force Treadmills: Validation of an Instrument , 2008, American journal of physical medicine & rehabilitation.

[12]  S. Gard,et al.  The influence of stance-phase knee flexion on the vertical displacement of the trunk during normal walking. , 1999, Archives of physical medicine and rehabilitation.

[13]  Massimo Piccardi,et al.  Height measurement as a session-based biometric for people matching across disjoint camera views , 2005 .

[14]  Antonio Criminisi,et al.  New approach to obtain height measurements from video , 1999, Other Conferences.

[15]  J. Perry,et al.  Gait Analysis , 2024 .

[16]  Yaser Yacoob,et al.  Statistical body height estimation from a single image , 2008, 2008 8th IEEE International Conference on Automatic Face & Gesture Recognition.

[17]  Remo Sala,et al.  New method for height estimation of subjects represented in photograms taken from video surveillance systems , 2007, International Journal of Legal Medicine.