Fast-r-cnn-pedestrian-detection-with-CAFFE-and-GPU-support

C++ and Python2.7 implementation of a automatic system for pedestrian detection at night using far infrared visual information based on convolutional neural networks.

Getting Started

The ProgramaTesis folder contains the implementation of the system in python 2.7 and the ProgramasTesisGPU/ProgramaTesisC++/ folder contains the implementation of the system in C ++ with GPU support.

To run the system you need to run either the file https://github.com/lmbarr/Fast-r-cnn-pedestrian-detection-with-CAFFE-and-GPU-support/blob/master/ProgramaTesis/ejecutar.py for python or the file https://github.com/lmbarr/Fast-r-cnn-pedestrian-detection-with-CAFFE-and-GPU-support/blob/master/ProgramasTesisGPU/ProgramaTesisC%2B%2B/main_videos.cpp for c++.

Prerequisites

The software was executed on:

• Ubuntu 14.04

using:

• opencv and opencv_contrib 3.1.0 http://www.pyimagesearch.com/2015/06/22/install-opencv-3-0-and-python-2-7-on-ubuntu/
• caffe-fast https://github.com/rbgirshick/fast-rcnn
• gcc 4.8.5
• python2 2.7.6

More Information:

See https://github.com/lmbarr/Fast-r-cnn-pedestrian-detection-with-CAFFE-and-GPU-support/blob/master/Manual%20de%20usuario.pdf

QA:

Let me know your questions at [email protected].