VLR Course Assignments

This repository contains completed assignments for the Visual Learning and Recognition (VLR) course, covering various computer vision and deep learning topics.

Overview

The repository contains three comprehensive assignments:

1. Assignment 1: Object Detection using FCOS (Fully Convolutional One-Stage Object Detection)

   • Implementation of the FCOS detector with FPN (Feature Pyramid Network)
   • Tasks include feature extraction, box regression, and centerness prediction
   • Code for training and testing the detector on the PASCAL VOC dataset

2. Assignment 2: Generative Models

   • GAN implementation with different loss functions (Original GAN, LSGAN, WGAN-GP)
   • VAE implementation with different latent space dimensions and beta values
   • Diffusion models with DDPM and DDIM sampling techniques

3. Assignment 3: Transformers for Vision

   • Transformer for image captioning using the COCO dataset
   • Vision Transformer (ViT) implementation for image classification

Additional Information

Please refer to the individual README files in each assignment directory for:

• Detailed setup instructions and dependencies
• Dataset preparation and requirements
• Implementation details
• Training and evaluation procedures