Skip to content

Latest commit

 

History

History
31 lines (21 loc) · 857 Bytes

README.md

File metadata and controls

31 lines (21 loc) · 857 Bytes

Person Detectors

A collection of person detectors used for person detection, tracking and related image processing research and applications. At the moment this repo contains person detectors:

  • Harr cascade based person detector
  • HOG based person detector
  • Blob person detector
  • Mask RCNN (Object detection + segmentation based) person detector
  • Tensorflow Object detection based person detector
  • OpenPose tensorflow implementation based person detector

The main objective of this repository is to allow comparisons to be done on different person detectors.

Example

python DetectorPreview.py <path to test video> <detector>

detector can be one of haar, blob, mask, op, tfod, tfop and hog.

Contributors

Many thanks to Madhawa Vidanapathirana for working on evaluating these detectors.