PfAS-Project

34759 Perception for Autonomous Systems - Final Project

Setup

Add the raw images in the raw_data folder, and rectified data in the rec_data folder.

python3.11 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

Content

• camera_calib.ipynb, to compute the camera calibration matrices, mainly written by Lkjaersgaard
• load.py, to properly load the data, mainly written by Seb-sti1
• depth.py, to compute the disparity/depth associated to the stereo images, mainly written by Seb-sti1
• viz.py, visualize the data (images, pcd...), mainly written by Seb-sti1
• detect.py, viz_yolo.py, compare the result of the YOLO with the truth, mainly written by nandortakacs
• train.py, train the YOLO model, mainly written by nandortakacs
• object_detection.py, try several clustering algorithm to detect relevant objects, mainly written by Seb-sti1
• kalman_filter_z.py and kalman_filter_2d.py, kalman filter implementation, mainly written by nandortakacs
• main.py, use the implemented functions in the other files to test to solve the whole problem, mainly written by Seb-sti1