People reidentification in surveillance and forensics

Abstract

The field of surveillance and forensics research is currently shifting focus and is now showing an ever increasing interest in the task of people reidentification. This is the task of assigning the same identifier to all instances of a particular individual captured in a series of images or videos, even after the occurrence of significant gaps over time or space. People reidentification can be a useful tool for people analysis in security as a data association method for long-term tracking in surveillance. However, current identification techniques being utilized present many difficulties and shortcomings. For instance, they rely solely on the exploitation of visual cues such as color, texture, and the object’s shape. Despite the many advances in this field, reidentification is still an open problem. This survey aims to tackle all the issues and challenging aspects of people reidentification while simultaneously describing the previously proposed solutions for the encountered problems. This begins with the first attempts of holistic descriptors and progresses to the more recently adopted 2D and 3D model-based approaches. The survey also includes an exhaustive treatise of all the aspects of people reidentification, including available datasets, evaluation metrics, and benchmarking.

Person Re-Identification Using Kernel-Based Metric Learning Methods

“…A good overview of existing re-ID methods can be found in [7,10,13,23,29] and references therein. The three most important aspects in re-ID are i) the features used, ii) the matching procedure, and iii) the performance evaluation.…”

Soft biometric retrieval to describe and identify surveillance images

“…We explore some of the questions raised in this paper, with regards to inter-and intra-class variation, small numbers of samples per subject, and generalising capability across cameras, using our relatively small, but highly detailed dataset of 100 subjects. Vezzani et al highlight the application benefits of measuring a biometric profile, enabling both reidentification and biometric recognition [27]. Finally, Wang et al present a strong argument for using semantic features, stating qualities such as: viewpoint independence, robustness and capacity for human interpretation [28].…”

A multimedia system to produce and deliver video fragments on demand on parliamentary websites

“…Thus, the division of the video into shots is the first step to be carried out by the detection tool. The shot boundary detection problem has traditionally involved the computation of visual differences between consecutive frames and a shot has been detected when this difference has been higher than a certain threshold [31,43,46]. In our approach, the feature used in each frame to detect the shot transition is the color, which has proven to be effective [47].…”