2009 Twelfth IEEE International Workshop on Performance Evaluation of Tracking and Surveillance 2009
DOI: 10.1109/pets-winter.2009.5399556
|View full text |Cite
|
Sign up to set email alerts
|

PETS2009: Dataset and challenge

Abstract: This paper describes the crowd image analysis challenge that forms part of the PETS 2009 workshop. The aim of this challenge is to use new or existing systems for i) crowd count and density estimation, ii) tracking of individual(s) within a crowd, and iii) detection of separate flows and specific crowd events, in a real-world environment. The dataset scenarios were filmed from multiple cameras and involve multiple actors.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
217
0
6

Year Published

2012
2012
2022
2022

Publication Types

Select...
5
2
2

Relationship

0
9

Authors

Journals

citations
Cited by 419 publications
(223 citation statements)
references
References 0 publications
0
217
0
6
Order By: Relevance
“…1. For our study, we use the first half (frames 1-218) of the challenging PETS S2.L2 dataset [1] as our mining sequence, and we employ a recent multi-target tracker based on energy minimization [2]. The ouput of the method is a joint detector that is tailored to detect occlusion patterns that are most relevant for multi-target tracking.…”
Section: Inputmentioning
confidence: 99%
See 1 more Smart Citation
“…1. For our study, we use the first half (frames 1-218) of the challenging PETS S2.L2 dataset [1] as our mining sequence, and we employ a recent multi-target tracker based on energy minimization [2]. The ouput of the method is a joint detector that is tailored to detect occlusion patterns that are most relevant for multi-target tracking.…”
Section: Inputmentioning
confidence: 99%
“…We evaluate the proposed joint person detector and its application to tracking on three challenging sequences: PETS S2.L2, S1.L2 [1], and ParkingLot [3]. By analyzing and mining occlusion patterns, we obtain very competitive detection results both in terms of recall and precision, as shown in Fig.…”
Section: Experiments and Conclusionmentioning
confidence: 99%
“…We demonstrate our approach on eight different sequences. The first set consists of six publicly available videos including the PETS 2009 benchmark [34] 1 and TUD Stadtmitte [35]. All videos show pedestrians in a single view but they exhibit a large variation in person count, camera viewpoint and motion patterns.…”
Section: Methodsmentioning
confidence: 99%
“…The challenging factors of these sequences are listed in Table 1. car [19] large illumination change, distraction from other objects jumping [13] image blur, fast motion face [1] long-duration occlusion singer [14] large illumination variation, large scale change PETS2009 [8] out-of-plane pose change, heavy occlusion Avatar1 large scale change, low contrast Construct an initial over-complete dictionary D 1 , and learn a linear classifier with parameter w 1 .…”
Section: Methodsmentioning
confidence: 99%