By Taihong Xiao and Jinwen Ma
This repo is an Matlab implementation of our paper. ILFPT is in integrated framework designed for pedestrian tracking, especially in the surveillance videos.
Here is a video demo.
Caffe
(seeexternal/caffe/
)- MATLAB
- GPU: GTX 980/1080, Tesla K20/K40/K80
git clone --recursive https://github.com/Prinsphield/ILFPT.git
- Build Caffe in the
external/caffe
- Download the test videos from either Google Drive or BaiduPan and extract them into the 'test/' directory
- Download the detection network model from either Google Drive or BaiduPan and extract into
models/
directory - Download our pretrained detection network from either Google Drive or BaiduPan and extract into
output/
directory - Download the pretrained re-id network files from either Google Drive or BaiduPan and extract into
reid_net/
directory
- Start Matlab from the root directory
- Run
faster_rcnn_build.m
- Run
startup.m
- Run
demo.m
- Download SVD-B training data from OneDrive. Extract them into
dataset/
directory and rename toVOCdevkit2007/
- Modify related files in
models/
dir to config detection network - Run scripts in
experiments/
accordingly to train a detection network.
Note that the GPU cost for training a detection network is much higher than that for testing. Before training your own detection network, please ensure that your GPU memory memory meets the following requirement:
- 3GB GPU memory for ZF net
- 9GB GPU memory for VGG-16 net