54 Days of Computer Vision.

54_Days_CV.md

Computer Vision and Deep Learning Topics by Difficulty Level

Easy

• FCN (Fully Convolutional Networks)
• UNet: Standard architecture for biomedical image segmentation
• YOLO Series: Real-time object detection family
• CAM (Class Activation Mapping)
• VGGNet
• SqueezeNet: Lightweight architecture
• EfficientNet: Scaling networks efficiently
• ResNet: Residual Networks fundamentals
• SSD (Single Shot Detector)
• Basic Attention Mechanisms
• Group Normalization
• Transfer Learning Basics

Medium

• Vision Transformer (ViT)
• DETR (Detection Transformer)
• RetinaNet
• Mask R-CNN
• FPN (Feature Pyramid Networks)
• Yolov5 and Advanced YOLO variants
• DeiT (Data-efficient image Transformer)
• Graph Convolutional Networks
• CenterNet
• RepVGG
• EfficientDet
• Focal Loss and Advanced Loss Functions
• Grad-CAM
• DeepLab Series
• Attention Mechanisms in Vision

Intermediate

• StyleGAN Series
• Swin Transformer
• CLIP (Contrastive Language-Image Pre-training)
• NeRF (Neural Radiance Fields)
• Advanced Transformer Architectures
• Graph Attention Networks
• Panoptic Segmentation
• Self-Attention and Multi-Head Attention
• Advanced Object Detection
• Instance Segmentation
• Few-Shot Learning
• Self-Supervised Learning
• Metric Learning

Hard

• Neural Architecture Search
• Advanced GAN Architectures
• Multi-Modal Learning
• 3D Vision Transformers
• Advanced NeRF Variants
• Graph Neural Networks Theory
• Meta-Learning
• Self-Supervised Representation Learning
• Advanced Optimization Techniques
• Vision-Language Models
• Quantum Computer Vision
• Neural ODEs
• Advanced Generative Models
• Theoretical Deep Learning

Note: Topics are categorized based on prerequisite knowledge, mathematical complexity, and implementation difficulty. Individual topics may span multiple difficulty levels depending on depth of study. To mark a topic first foork this list and the mark them as completed, replace "[ ]" with "[x]" in the markdown.