Pedestrian Detection and Tracking

Instructor : Dr. Amitabha Mukerjee

Group V6 : Vimal Sharma, Shikhar Sharma

Abstract

We use deformable part based models [3] of human body to detect[2] and track pedestrians in a video. These models must be robust and capable of detecting pedestrians in a wide variety of poses/clothing and even if some of their body parts are occluded.

Project Report

Click here

Code and Datasets

Code : voc-release4.01.tgz
Code : Mean Shift
Dataset : PASCAL VOC 2006 Development Kit, PASCAL VOC 2009 Development Kit
Dataset : Weizmann Classification Database
Our Modified Code Files : Tarball, Readme

Files used for calculating results: Files

Results

Sample Detections of [1]

Tracking using Background Subtraction

Tracking using Mean Shift

We picked up random google images obtained on searching "people", "hu- mans", "people walking", counted the number of people(occluded and unoc- cluded seperately) and then counted the number of detections. The results are summarized in this table:

Type	Total Number	Detected	Percentage
All	134	97	72.4
Unoccluded	61	55	90.2
Occluded	73	42	57.6

References

[1] Pedro F. Felzenszwalb, Ross B. Girshick, David McAllester and Deva Ramanan[2010] Object Detection with Discriminatively Trained Part-Based Models. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. 32, NO. 9, SEPTEMBER 2010

[2] P. F. Felzenszwalb, R. B. Girshick, and D. McAllester. Discriminatively trained deformable part models, release 4. http://people.cs.uchicago.edu/ pff/latent-release4/.

[3] Pedro F. Felzenszwalb and Daniel P. Huttenlocher. Pictorial struc- tures for object recognition. International Journal of Computer Vision, 61(1):55-79, 2005.

[4] Marcelo Molina Adam Kukucka, Zach Clay. Implementation of mean shift tracker. http://www.hackchina.com/en/cont/17222.

[5] Navneet Dalal and Bill Triggs. Histograms of oriented gradients for human detection. In In CVPR, pages 886-893, 2005.

[6] M. Everingham, L. Van Gool, C. K. I. Williams, J. Winn, and A. Zis- serman. The pascal visual object classes (voc) challenge. International Journal of Computer Vision, 88(2):303-338, June 2010.

[7] Pedro F. Felzenszwalb. Object detection grammars. In ICCV Work- shops, page 691. IEEE, 2011.

[8] Deva Ramanan. Learning to parse images of articulated bodies. In In Proc. NIPS, 2006.

[9] Felix Hageloh Roberto Valenti. Implementation and evaluation of the mean shift tracker. http://staff.science.uva.nl/~rvalenti.

[10] P. Viola and M. Jones. Rapid object detection using a boosted cascade of simple features. In Computer Vision and Pattern Recognition, 2001. CVPR 2001. Proceedings of the 2001 IEEE Computer Society Conference on, volume 1, pages I-511 - I-518 vol.1, 2001.