Direction-based stochastic matching for pedestrian recognition in non-overlapping cameras

Chen, Xiaotang; Huang, Kaiqi; Tan, Tieniu

doi:10.1109/icip.2011.6115887

Cited by 11 publications

(9 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Lighting variations are addressed through color normalization [18], exemplar based approaches [20], or brightness transfer functions learned with [23,25] or without supervision [13,19,24,29]. Discriminative power is improved by saliency information [33,34] or learning features specific to body parts [6,18,[20][21][22][23]27], either in the image [35][36][37] or back-projected onto an articulated [38,39] or monolithic [40] 3D body model. All MTMC trackers employ optimization to maximize the coherence of observations for predicted identities.…”

Section: Related Workmentioning

confidence: 99%

Performance Measures and a Data Set for Multi-target, Multi-camera Tracking

Ristani

Solera

Zou

et al. 2016

Lecture Notes in Computer Science

2,054

1,467

View full text Add to dashboard Cite

Abstract. To help accelerate progress in multi-target, multi-camera tracking systems, we present (i) a new pair of precision-recall measures of performance that treats errors of all types uniformly and emphasizes correct identification over sources of error; (ii) the largest fully-annotated and calibrated data set to date with more than 2 million frames of 1080p, 60fps video taken by 8 cameras observing more than 2,700 identities over 85 minutes; and (iii) a reference software system as a comparison baseline. We show that (i) our measures properly account for bottom-line identity match performance in the multi-camera setting; (ii) our data set poses realistic challenges to current trackers; and (iii) the performance of our system is comparable to the state of the art.

show abstract

Section: Related Workmentioning

confidence: 99%

Performance Measures and a Data Set for Multi-target, Multi-camera Tracking

Ristani

Solera

Zou

et al. 2016

Lecture Notes in Computer Science

2,054

1,467

View full text Add to dashboard Cite

show abstract

“…[6,255,260,348,358]. Many objects trackers, like [37] rely on extracting appearance information from the object pixels using hand-crafted features, where including appearance information via colour histogram [38,47,49,50,80,192,317,329,367] and texture descriptors [38,49,79,187,373,374] are the most popular representation for appearance modelling in MOT.…”

Section: B) Appearance Modelsmentioning

confidence: 99%

“…50 Comparison of Multi-Object Tracking performance of MOT.2 with top 3 online benchmarking algorithms over MOT17 dataset.…”

mentioning

confidence: 99%

Multi-Object Tracking with Data Association by a Similarity Identification Model

Gómez-Silva

Armingol

Escalera

2016

7th International Conference on Imaging for Crime Detection and Prevention (ICDP 2016)

View full text Add to dashboard Cite

show abstract

“…[3] proposed an optimization framework by combining short term feature correspondences across the cameras with the long-term feature dependency models. [19] did not use the spatio-temporal cues in multi camera scenarios, but instead investigated directional angles using the spatio-temporal continuity in a single camera field. However, most of these works failed to handle a high clutter scene.…”

Section: Related Workmentioning

confidence: 99%

Tracking multiple interacting targets in a camera network

Zhang

Zhu

Roy-Chowdhury

2015

Computer Vision and Image Understanding

View full text Add to dashboard Cite

In this paper we propose a framework for tracking multiple interacting targets in a wide-area camera network consisting of both overlapping and non-overlapping cameras. Our method is motivated from observations that both individuals and groups of targets interact with each other in natural scenes. We associate each raw target trajectory (i.e., a tracklet) with a group state, which indicates if the trajectory belongs to an individual or a group. Structural support vector machine (SSVM) is applied to the group states to decide if merge or split events occur in the scene. Information fusion between multiple overlapping cameras is handled using a homography-based voting scheme. The problem of tracking multiple interacting targets is then converted to a network flow problem, for which the solution can be obtained by the K-shortest paths algorithm. We demonstrate the effectiveness of the proposed algorithm on the challenging VideoWeb dataset in which a large amount of multi-person interaction activities are present. Comparative analysis with state-of-the-art methods is also shown.

show abstract

Direction-based stochastic matching for pedestrian recognition in non-overlapping cameras

Cited by 11 publications

References 7 publications

Performance Measures and a Data Set for Multi-target, Multi-camera Tracking

Performance Measures and a Data Set for Multi-target, Multi-camera Tracking

Multi-Object Tracking with Data Association by a Similarity Identification Model

Tracking multiple interacting targets in a camera network

Contact Info

Product

Resources

About