A Data Association Algorithm for People Re-identification in Photo Sequences

Color-Spatial Person Re-identification by a Voting Matching Scheme

“…For example, Presti et al [17] presented a probabilistic framework to determine the match between a set of identities and observations. The probability density function was formulated as an exponential distribution of the Euclidean distance of descriptors.…”

An on-line learning method for face association in personal photo collection

“…In previous works [18,14], associations across a photo sequence are found by considering face and clothing features locally within a temporal window and by means of a joint probabilistic data association (JPDA) [19]. Basically, associations between identities and depictions in a photo are computed as maximum matching in a bipartite graph [20] by means of the Hungarian algorithm [21].…”

Automatic Generation of Subject-Based Image Transitions

“…Here we used a data association framework for people re-identification [1] that takes advantage of an important constraint: a person can not be present two times in the same photo and if a face is associated to an identity, the remaining faces in the same photo must be associated to other identities. The problem is modeled as the search for probable associations between faces detected in subsequent photos using face and clothing descriptions.…”

“…Starting from a collection of personal images, we propose a tool for the automatic slideshow of a sequence of pictures depicting the same individual. Firstly the whole collection is searched for faces, while face identities are assigned with an automatic approach [1]. Then each face is processed for automatically finding the position of some facial feature points.…”