PMID- 23428432
OWN - NLM
STAT- MEDLINE
DCOM- 20131104
LR  - 20130222
IS  - 1939-3539 (Electronic)
IS  - 0098-5589 (Linking)
VI  - 35
IP  - 4
DP  - 2013 Apr
TI  - Multiple target tracking by learning-based hierarchical association of detection 
      responses.
PG  - 898-910
LID - 10.1109/TPAMI.2012.159 [doi]
AB  - We propose a hierarchical association approach to multiple target tracking from a 
      single camera by progressively linking detection responses into longer track 
      fragments (i.e., tracklets). Given frame-by-frame detection results, a 
      conservative dual-threshold method that only links very similar detection 
      responses between consecutive frames is adopted to generate initial tracklets 
      with minimum identity switches. Further association of these highly fragmented 
      tracklets at each level of the hierarchy is formulated as a Maximum A Posteriori 
      (MAP) problem that considers initialization, termination, and transition of 
      tracklets as well as the possibility of them being false alarms, which can be 
      efficiently computed by the Hungarian algorithm. The tracklet affinity model, 
      which measures the likelihood of two tracklets belonging to the same target, is a 
      linear combination of automatically learned weak nonparametric models upon 
      various features, which is distinct from most of previous work that relies on 
      heuristic selection of parametric models and manual tuning of their parameters. 
      For this purpose, we develop a novel bag ranking method and train the crucial 
      tracklet affinity models by the boosting algorithm. This bag ranking method 
      utilizes the soft max function to relax the oversufficient objective function 
      used by the conventional instance ranking method. It provides a tighter upper 
      bound of empirical errors in distinguishing correct associations from the 
      incorrect ones, and thus yields more accurate tracklet affinity models for the 
      tracklet association problem. We apply this approach to the challenging multiple 
      pedestrian tracking task. Systematic experiments conducted on two real-life 
      datasets show that the proposed approach outperforms previous state-of-the-art 
      algorithms in terms of tracking accuracy, in particular, considerably reducing 
      fragmentations and identity switches.
FAU - Huang, Chang
AU  - Huang C
AD  - NEC Research Laboratories, Cupertino, CA 95014, USA. huangchang@baidu.com
FAU - Li, Yuan
AU  - Li Y
FAU - Nevatia, Ramakant
AU  - Nevatia R
LA  - eng
PT  - Journal Article
PL  - United States
TA  - IEEE Trans Pattern Anal Mach Intell
JT  - IEEE transactions on pattern analysis and machine intelligence
JID - 9885960
SB  - IM
MH  - *Algorithms
MH  - *Artificial Intelligence
MH  - Databases, Factual
MH  - Human Activities/classification
MH  - Humans
MH  - Image Processing, Computer-Assisted/*methods
MH  - Models, Theoretical
MH  - Movement
MH  - Pattern Recognition, Automated/*methods
MH  - ROC Curve
MH  - Video Recording/*methods
EDAT- 2013/02/23 06:00
MHDA- 2013/11/05 06:00
CRDT- 2013/02/23 06:00
PHST- 2013/02/23 06:00 [entrez]
PHST- 2013/02/23 06:00 [pubmed]
PHST- 2013/11/05 06:00 [medline]
AID - 10.1109/TPAMI.2012.159 [doi]
PST - ppublish
SO  - IEEE Trans Pattern Anal Mach Intell. 2013 Apr;35(4):898-910. doi: 
      10.1109/TPAMI.2012.159.