From: Identifying Persons of Interest in CCtV Camera
by Furqan M. Khan and Francois Bremond (Inria) in ERCIM News 108
"Person re-identification is a challenging task because individuals move in all directions, including away from the camera and across the field of view (see Figure 1). Therefore, biometric cues such as face or iris cannot be reliably extracted. Instead, holistic appearance of the person (clothing) or gait is used, which is inherently not as discriminative. Furthermore, due to low resolution, subtleties in gait are difficult to measure. In addition, a person’s appearance in a video is susceptible to illumination, occlusion, camera properties and viewing angle. Finally, for a fully automated system, individuals must be localised using a person detection and tracking algorithm before building their appearance models." p.35.
The above text prompted the musings below as per Hodges' model and a nursing - healthcare context:
Patient (Person) re-identification is a challenging task because patients move through the health and social care systems in many directions, routes and disciplines, including away from the humanistic values that should (must) inform practice for comprehensive assessment (see Figure 1 - Hodges' model). Therefore, mechanistic (that includes biometric) cues AND humanistic perspectives cannot be reliably extracted and integrated. Instead, holistic appearance of the person across four care or knowledge domains can be used. This is inherently not as discriminative as it starts with a blank (neutral) conceptual framework. Furthermore, emergency situations (hopefully) excepted, staffing pressures and time constraints create low resolution accounts and records in terms of person-centeredness and so subtleties in a variety of care strengths and needs (including gait) may be difficult to discern and measure. In addition, a person’s representation in a care episode or encounter is susceptible to communication problems (noise), lack of access to records, whether paper or computer-based, all aggravated by a viewing angle that is still service-centered. Finally, a system that can facilitate holistic, person centered and integrated care for individuals must be self-referential in that it facilitates definition of health and what care (by self, other person [or thing - robot]) encompasses. The system will encourage reflection and critical thinking helping to assure person-centered skills in staff ('person' detection) and continuity of care (tracking algorithm) before building the appearance model: their health career.