ChokePoint Dataset

Overview · Example · Protocol · Licence · Download · Contacts · Related Datasets · Acknowledgement


We collected a video dataset, termed ChokePoint, designed for experiments in person identification/verification under real-world surveillance conditions using existing technologies. An array of three cameras was placed above several portals (natural choke points in terms of pedestrian traffic) to capture subjects walking through each portal in a natural way (see example). While a person is walking through a portal, a sequence of face images (ie. a face set) can be captured. Faces in such sets will have variations in terms of illumination conditions, pose, sharpness, as well as misalignment due to automatic face localisation/detection. Due to the three camera configuration, one of the cameras is likely to capture a face set where a subset of the faces is near-frontal.

The dataset consists of 25 subjects (19 male and 6 female) in portal 1 and 29 subjects (23 male and 6 female) in portal 2. The recording of portal 1 and portal 2 are one month apart. The dataset has frame rate of 30 fps and the image resolution is 800X600 pixels. In total, the dataset consists of 48 video sequences and 64,204 face images. In all sequences, only one subject is presented in the image at a time. The first 100 frames of each sequence are for background modelling where no foreground objects were presented.

Each sequence was named according to the recording conditions (eg. P2E_S1_C3) where P, S, and C stand for portal, sequence and camera, respectively. E and L indicate subjects either entering or leaving the portal. The numbers indicate the respective portal, sequence and camera label. For example, P2L_S1_C3 indicates that the recording was done in Portal 2, with people leaving the portal, and captured by camera 3 in the first recorded sequence.

To pose a more challenging real-world surveillance problems, two seqeunces (P2E_S5 and P2L_S5) were recorded with crowded scenario. In additional to the aforementioned variations, the sequences were presented with continuous occlusion. This phenomenon presents challenges in identidy tracking and face verification.

This dataset can be applied, but not limited, to the following research areas:

  • person re-identification
  • image set matching
  • face quality measurement
  • face clustering
  • 3D face reconstruction
  • pedestrian/face tracking
  • background estimation and substraction


An example of the recording setup used for the ChokePoint dataset. A camera rig contains 3 cameras placed just above a door, used for simultaneously recording the entry of a person from 3 viewpoints. The variations between viewpoints allow for variations in walking directions, facilitating the capture of a near-frontal face by one of the cameras.

Camera Rig


Camera 1

Camera 2

Camera 3

Example shots from the ChokePoint dataset, showing portals with various backgrounds.

P1E - Camera 1

P1L - Camera 1

P2E - Camera 2

P2L - Camera 2

P2E_S5 - Camera 2

P2L_S5 - Camera 2


We designed a baseline verification protocol (protocol_baseline) for this dataset. In this protocol, video sequences are divided into two groups (G1 and G2), where each group played the role of development set and evaluation set in turn. In each group, all possible genuine and imposter pairs were generated. Parameters and threshold are first learned on the development set and then applied on the evaluation set. The average verification rate is used for reporting results.


To study the effect of person verification under different environments and time interval between recording, following case studies can be considered:

  1. case_study_1
    • indoor scene only
    • short time interval


  2. case_study_2
    • indoor and outdoor scene
    • short time interval


  3. case_study_3
    • indoor and outdoor scene
    • long time interval
    • genuine and imposter pairs shall be generated by matching each sequence of set 1 with set 2.

    Set 1 Set 2
    G1 P1E_S1_C1 P1E_S2_C2 P1L_S1_C1 P1L_S2_C2
    P2E_S2_C2 P2E_S1_C3 P2L_S2_C2 P2L_S1_C1
    G2 P1E_S3_C3 P1E_S4_C1 P1L_S3_C3 P1L_S4_C1
    P2E_S4_C2 P2E_S3_C1 P2L_S4_C2 P2L_S3_C3

We also encourage the experiments to be conducted with two evaluation conditions:

  1. single camera (SC)
    - using faces from the camera with most frontal view (listed in above tables).
  2. multiple camera (MC)
    - using faces from all three cameras. For example, images for portal 1E and sequence 1 will are taken from P1E_S1_C1, P1E_S1_C2 and P1E_S1_C3.


This dataset ('Licensed Material') is made available to the scientific community for non-commercial research purposes such as academic research, teaching, scientific publications or personal experimentation. Permission is granted by National ICT Australia Limited (NICTA) to you (the 'Licensee') to use, copy and distribute the Licensed Material in accordance with the following terms and conditions:

  1. Licensee must include a reference to NICTA and the following publication in any published work that makes use of the Licensed Material:

      Y. Wong, S. Chen, S. Mau, C. Sanderson, B.C. Lovell
      Patch-based Probabilistic Image Quality Assessment for Face Selection and Improved Video-based Face Recognition
      IEEE Biometrics Workshop, Computer Vision and Pattern Recognition (CVPR) Workshops, pages 81-88. IEEE, June 2011.

      Bibtex entry:
         AUTHOR    = {Yongkang Wong and Shaokang Chen and Sandra Mau and Conrad Sanderson and Brian C. Lovell},
         TITLE     = {Patch-based Probabilistic Image Quality Assessment for Face Selection and Improved Video-based Face Recognition},
         BOOKTITLE = {IEEE Biometrics Workshop, Computer Vision and Pattern Recognition (CVPR) Workshops},
         YEAR      = {2011},
         pages     = {81-88},
         month     = {June},
         publisher = {IEEE}

  2. If Licensee alters the content of the Licensed Material or creates any derivative work, Licensee must include in the altered Licensed Material or derivative work prominent notices to ensure that any recipients know that they are not receiving the original Licensed Material.

  3. Licensee may not use or distribute the Licensed Material or any derivative work for commercial purposes including but not limited to, licensing or selling the Licensed Material or using the Licensed Material for commercial gain.

  4. The Licensed Material is provided 'AS IS', without any express or implied warranties. NICTA does not accept any responsibility for errors or omissions in the Licensed Material.

  5. This original license notice must be retained in all copies or derivatives of the Licensed Material.

  6. All rights not expressly granted to the Licensee are reserved by NICTA.



  • The ChokePoint dataset taking up about 12 Gb. Each tar.xz file is on average around 200 Mb

  • Brief description of the data can be found in README

  • The cropped face images were extracted using the manually labelled eye location. The faces has the size of 96x96 pixels

  • Please download only one file at a time -- this is so the server is not overloaded

  • Microsoft Windows user can extract the *.tar.xz files with 7-Zip

Original files:

  1. groundtruth.tar.xz
  2. P1E_S1.tar.xz
  3. P1E_S2.tar.xz
  4. P1E_S3.tar.xz
  5. P1E_S4.tar.xz
  6. P1L_S1.tar.xz
  7. P1L_S2.tar.xz
  8. P1L_S3.tar.xz
  9. P1L_S4.tar.xz
  10. P2E_S1.tar.xz
  11. P2E_S2.tar.xz
  12. P2E_S3.tar.xz
  13. P2E_S4.tar.xz
  14. P2E_S5.tar.xz
  15. P2L_S1.tar.xz
  16. P2L_S2.tar.xz
  17. P2L_S3.tar.xz
  18. P2L_S4.tar.xz
  19. P2L_S5.tar.xz

Cropped face images:

  1. P1E.tar.xz
  2. P1L.tar.xz
  3. P2E.tar.xz
  4. P2L.tar.xz


If you have any questions regarding to the dataset, please contact:

    yongkang döt wong ät ieee döt orgxyz

Related Datasets

  • VB100 Bird Dataset - surveillance videos of 100 bird species, for experiments in fine-grained classification
  • VidTIMIT Dataset - audio-visual recordings of people reciting phonetically balanced sentences


The chokepoint dataset is sponsored by NICTA. NICTA is funded by the Australian Government as represented by the Department of Broadband, Communications and the Digital Economy, as well as the Australian Research Council through the ICT Centre of Excellence program.