This project was for the course CS-271: Topics in Machine Learning.

Multi-class Object Detection in Images

The objective of this project was to detect 5 different classes - car, truck, pedestrian, traffic light and bicyclist - in a given image. The task of object detection was be done using 2 different machine learning algorithms; SVM and YOLO algorithm.

Dataset

The dataset that was used for this project was the Self-Driving Cars dataset on Kaggle. This dataset contains 22,241 images, divided into train and test splits, where the images contain multiple objects from the given 5 classes.

Report

A report, containing all the information and results related to this project, is available in the report.pdf file.

Code Files

The code files related to training the models can be found in the following locations:

• svm.ipynb
• yolov5.ipynb