Caltech Pedestrian to YOLO Converter

This repository provides a set of tools to prepare Caltech Pedestrian dataset to the format of YOLO object detector. The toolbox contains three main modules for preparing Caltech Pedestrian data for different versions of YOLO, described as below:

• Image Generator: Generates a set of .png images from Caltech Pedestrian .seq files. By feeding the root directory that contains set00X folders in the config file, the process of generating images can start.
• Annotation Generator: Generates a set of .txt annotation (label) files from Caltech Pedestrian .vbb files.
• Plot Annotations: Draws bounding boxes using annotations on sample generated images. The output will be a video file showing the consequent frames and drawn labels. You need to run this module after the two above.

Please note that set00-set05 are training data and the rest (i.e., set06-set10) are test data (see link).

⚙️ Prerequisites

You will need below libraries before running the application.

• Python >= 3.7
• Numpy >= 1.19
• Scipy >= 1.6
• PyInquirer >= 1.0.3
• Opencv-python >= 4.1.1

As an alternative, simply run the below command (root directory):

pip install -r requirements.txt

🚀 Launch the Application

The first step to run the engine of the application is to provide a proper configuration file. Accordingly, make a copy of the config.example.py file in the root directory and rename it to config.py. Comments can help you to define proper configurations. Then, you can simply run the program by:

python ./main.py

💡 Inspiration

• The framework of Shunta Saito
• A similar work by Simonzachau