Learning Implicit Transfer for Person Re-identification

Avraham, Tamar; Gurvich, Ilya; Lindenbaum, Michael; Markovitch, Shaul

doi:10.1007/978-3-642-33863-2_38

Cited by 84 publications

(104 citation statements)

References 11 publications

Supporting

Mentioning

102

Contrasting

Order By: Relevance

“…Re-identification: There is now an extensive body of research on conventional re-identification, broadly split into contributing effective feature representations [8,34], discriminative matching models [3,1], or both [19]. The other major design axis typically considered is 'single-shot' [3,1,34] (exactly one image per person) versus 'multi-shot' [8,16,27] (exploiting multiple images per person where available to improve results).…”

Section: Related Workmentioning

confidence: 99%

“…The other major design axis typically considered is 'single-shot' [3,1,34] (exactly one image per person) versus 'multi-shot' [8,16,27] (exploiting multiple images per person where available to improve results). For a broad background of research to this paper we suggest [11] and [32].…”

Section: Related Workmentioning

confidence: 99%

“…In contrast, the MRP re-id scenario is intrinsically open-world: self movement in a potentially open-space means one-to-one match situations are unlikely, self-motion means that tracking cannot provide multi-shot grouping, and clearly the person count of an arbitrarily surveilled space is not known in advance. Generalised-view Re-Id: The conventional approach to maximising re-id performance is learning a discriminative model to maximise re-id rate for a specific pair of fixed camera views [3,1]. A few studies have started to consider how re-identification models generalise across views [20] and generally found that they don't -achieving good re-id rate requires view specific discriminative training.…”

Section: Related Workmentioning

confidence: 99%

“…We investigate this by training a selection of strong discriminative models including one of the most popular: RankSVM [25]; and two recent state of the art approaches BR-SVM [2] and KISS [18]. We train these models on a variety of large benchmark datasets including VIPER [28] (632 distinct persons in [128x48] crops), PRID [13] (200 distinct persons), GRID [22] (250 persons) and CUHK [21] (971 persons).…”

Section: Classifier Training Representation and Datasetsmentioning

confidence: 99%

“…Binary-relation SVM with RBF concatenation kernel [2]. RankSVM-[DA] SVM with difference feature and linear kernel [25].…”

Section: Br-svm-[da]mentioning

confidence: 99%

See 4 more Smart Citations

Investigating Open-World Person Re-identification Using a Drone

Layne

Hospedales

Gong

2015

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Person re-identification is now one of the most topical and intensively studied problems in computer vision due to its challenging nature and its critical role in underpinning many multi-camera surveillance tasks. A fundamental assumption in almost all existing re-identification research is that cameras are in fixed emplacements, allowing the explicit modelling of camera and inter-camera properties in order to improve re-identification. In this paper, we present an introductory study pushing re-identification in a different direction: re-identification on a mobile platform, such as a drone. We formalise some variants of the standard formulation for re-identification that are more relevant for mobile reidentification. We introduce the first dataset for mobile re-identification, and we use this to elucidate the unique challenges of mobile re-identification. Finally, we re-evaluate some conventional wisdom about re-id models in the light of these challenges and suggest future avenues for research in this area.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Classifier Training Representation and Datasetsmentioning

confidence: 99%

“…Binary-relation SVM with RBF concatenation kernel [2]. RankSVM-[DA] SVM with difference feature and linear kernel [25].…”

Section: Br-svm-[da]mentioning

confidence: 99%

See 3 more Smart Citations

Investigating Open-World Person Re-identification Using a Drone

Layne

Hospedales

Gong

2015

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

The Re-identification Challenge

Gong

Cristani

Loy

et al. 2014

Person Re-Identification

126

102

View full text Add to dashboard Cite

Evaluating Feature Importance for Re-identification

Liu

Gong

Loy

et al. 2014

Person Re-Identification

View full text Add to dashboard Cite

Person re-identification methods seek robust person matching through combining feature types. Often, these features are assigned implicitly with a single vector of global weights, which are assumed to be universally and equally good for matching all individuals, independent of their different appearances. In this study, we present a comprehensive comparison and evaluation of up-to-date imagery features for person re-identification. We show that certain features play more important roles than others for different people. To that end, we introduce an unsupervised approach to learning a bottom-up measurement of feature importance. This is achieved through first automatically grouping individuals with similar appearance characteristics into different prototypes/clusters. Different features extracted from different individuals are then automatically weighted adaptively driven by their inherent appearance characteristics defined by the associated prototype. We show comparative evaluation on the re-identification effectiveness of the proposed prototype-sensitive feature importance-based method as compared to two generic weight-based global feature importance methods. We conclude by showing that their combination is able to yield more accurate person re-identification.

show abstract

Learning Implicit Transfer for Person Re-identification

Cited by 84 publications

References 11 publications

Investigating Open-World Person Re-identification Using a Drone

Investigating Open-World Person Re-identification Using a Drone

The Re-identification Challenge

Evaluating Feature Importance for Re-identification

Contact Info

Product

Resources

About