Person Re-identification for Improved Multi-person Multi-camera Tracking by Continuous Entity Association

@article{Narayan2017PersonRF,
  title={Person Re-identification for Improved Multi-person Multi-camera Tracking by Continuous Entity Association},
  author={Neeti Narayan and Nishant Sankaran and Devansh Arpit and Karthik Dantu and Srirangaraj Setlur and Venu Govindaraju},
  journal={2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)},
  year={2017},
  pages={566-572}
}
We present a novel approach to person tracking within the context of entity association. In large-scale distributed multi-camera systems, person re-identification is a challenging computer vision task as the problem is two-fold: detecting entities through identification and recognition techniques; and connecting entities temporally by associating them in often crowded environments. Since tracking essentially involves linking detections, we can reformulate it purely as a re-identification task… CONTINUE READING

Similar Papers

Results and Topics from this paper.

Key Quantitative Results

  • ● Evaluated performances of individual features for tracking achieving AUC scores of: ○ Face features – 96.56% ○ Attribute features – 99.37% ○ Location transition – 98.28% 26 Results DukeMTMC ● Evaluated performances of individual features for tracking achieving AUC scores of: ○ Face features – 92.07 ○ Attribute features – 99.99% ○ Location transition – 98.73% 27 Inference Results ● Inference error rates using proposed entity association algorithm ○ CamNeT: ■ Face features – 4.67% ■ Attribute features – 2.9% ■ Location transition– 4.49% ○ DukeMTMC: ■ Face features – 12.07% ■ Attribute features – 0.01% ■ Location transition– 0.5% 28 Comparison ● CamNet dataset: ○ Crossing fragments (XFrag): The number of true associations missed by the tracking system.

Citations

Publications citing this paper.
SHOWING 1-3 OF 3 CITATIONS

Re-identification for Online Person Tracking by Modeling Space-Time Continuum

  • 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
  • 2018
VIEW 4 EXCERPTS
CITES BACKGROUND, METHODS & RESULTS
HIGHLY INFLUENCED

State-aware Re-identification Feature for Multi-target Multi-camera Tracking

Peng Li, Jiabin Zhang, +3 authors Guan Huang
  • ArXiv
  • 2019
VIEW 1 EXCERPT
CITES BACKGROUND

References

Publications referenced by this paper.
SHOWING 1-10 OF 33 REFERENCES

A Camera Network Tracking (CamNeT) Dataset and Performance Baseline

  • 2015 IEEE Winter Conference on Applications of Computer Vision
  • 2015
VIEW 4 EXCERPTS
HIGHLY INFLUENTIAL

Social LSTM: Human Trajectory Prediction in Crowded Spaces

  • 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2016

FlowNet: Learning Optical Flow with Convolutional Networks

  • 2015 IEEE International Conference on Computer Vision (ICCV)
  • 2015
VIEW 1 EXCERPT

Learning to rank in person re-identification with metric ensembles

  • 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2015
VIEW 1 EXCERPT

Near-Online Multi-target Tracking with Aggregated Local Flow Descriptor

  • 2015 IEEE International Conference on Computer Vision (ICCV)
  • 2015
VIEW 1 EXCERPT

Person Re-Identification by Iterative Re-Weighted Sparse Ranking

  • IEEE Transactions on Pattern Analysis and Machine Intelligence
  • 2015
VIEW 1 EXCERPT