Day 01: Introduction to AI and Computer Vision & Image Classification:

• Definition of the problem and review of CNN architecture
• Common CNN architectures: AlexNet → VGG → ResNet → ResNext → GoogleNet → EfficientNet
• Deep understanding of CNN: filter visualization, feature visualization, activation map

Day 02: Object Detection:

• Definition of the problem and some methods prior to Deep Learning
• Introduction to the R-CNN family, YOLO (v1-v10), and SSD models
• Comparison and analysis of the advantages and disadvantages of the models, evaluation metrics

Day 03: Image Segmentation:

• Introduction to the problem and some algorithms prior to Deep Learning
• Algorithms: UNet, Mask R-CNN, DeepLab v1, v2, v3, SAM

Day 04: Time-Series Model:

• Introduction to the problem
• RNN, LSTM, GRU, Transformer, Vision Transformer
• Applications: 
  • Video Classification:
    • CNN-RNN
    • Transformer
  • Next-Frame Prediction: Convolutional LSTMs

Day 05: Image-Text and Image-Image problems:

• Introduction to Image Captioning, Visual Question Answering, Image Generation
• Image Captioning: LSTM, Attention, RNN, CNN, LSTM, CNN, CNN, RNN, LSTM
• Visual Question Answering: CNN, LSTM, CNN/RNN, L2Norm, GRU, LSTM, RNN, ResNet, Attention, MUTAN
• Image2Image: StyleGAN, PatchGAN, StarGAN, LDM, PCA, DDIM
• Txt2Image: LEDITS++, SINE, EditBench, MirrorGAN, DiffRF

Day 06: Some advanced problems:

• Introduction to OCR, Face Recognition, Pose Estimation/Gait Recognition, Object Tracking
• OCR: CRAFT, SRCNN
• Face Recognition: Linear Regression (LRC), DEF, CNN, GMM, VGG, GoogleNet
• Gait Recognition: GaitSet, OpenGait, CNN
• Pose Estimation: CNN, SVM, DNN, FPD, CNN, CRF
• Object Tracking: SOTMOT, MeMOT, CNN, Transformer

Day 07: Applications development:

• Web: Flask, FastAPI, Gradio, Ngrok
• Code Version Control: Github, Gitlab

Day 08: Deployment:

• Cloud: AWS EC2, GG Cloud
• Web Server: Nginx, Apache

Day 09: Automation:

• Docker, CI/CD

GRADUATION PROJECTS