Gaussian_yolo_HSV_traffic_classification

Using Gaussian Yolo in object detection and classifying Traffic_lights using HSV

Getting Started

These instructions will get you a copy of the project up and running on your local machine for development and testing purposes.

Prerequisites

Needed libraries

torch==1.0.0

numpy==1.15.2

matplotlib==3.0.2

opencv_python==3.4.4.19

tensorboardX==1.4

PyYAML>=4.2b1

pycocotools==2.0.0

seaborn==0.9.0

scikit-image

pip install -r requirements/requirements.txt

Download Gaussian Yolov3 Weights for ms-coco Dataset

FILE_ID="1zAFDSga9XLrsUBNHV3S2SvL1YWEsDB_p"
curl -sc /tmp/cookie "https://drive.google.com/uc?export=download&id=${FILE_ID}" > /dev/null
CODE="$(awk '/_warning_/ {print $NF}' /tmp/cookie)"
curl -sLb /tmp/cookie "https://drive.google.com/uc?export=download&confirm=${CODE}&id=${FILE_ID}" -o gaussian_yolov3_coco.pth 

Download MS-COCO Dataset

bash requirements/getcoco.sh

Running the scripts

Evaluation on COCO

python train.py --cfg config/gaussian_yolov3_eval.cfg --eval_interval 1  --checkpoint gaussian_yolov3_coco.pth

2D detection

Images

python run.py

Videos

python video.py

Stereo Vision

Images

python run_stereo.py

Video

python video_stereo.py

Generating Plots

python graphs.py