Month 4 — Computer Vision + NLP
🎯 Goal of this month
By the end of the month you will be able to:
- Classical image processing with OpenCV
- Object detection with YOLO and other pretrained models
- Text extraction with OCR (Tesseract, EasyOCR, PaddleOCR)
- Text preprocessing with spaCy and NLTK
- Applying BERT-style models with HuggingFace Transformers
Weekly breakdown
| Week | Topic | Time |
|---|---|---|
| Week 1 | OpenCV + Image Processing | 10-12 hours |
| Week 2 | YOLO, Detection, Segmentation, OCR | 10-12 hours |
| Week 3 | NLP foundations + Text Preprocessing | 10-12 hours |
| Week 4 | Transformers + HuggingFace | 10-12 hours |
Chapter order
- Introduction to Computer Vision
- Working with OpenCV
- YOLO and Object Detection
- NLP foundations
- Text Preprocessing
- Introduction to Transformers
- Exercises
What can you do by the end of the month?
- FastAPI service that accepts image/video upload and returns YOLO object detection
- OCR service — extracting data from passports and ID cards
- Sentiment analyzer and NER with HuggingFace pretrained models
- Be fully ready for Month 5 (LLM/RAG)
Tip for Backend Dev
This month mostly uses pretrained models:
- ResNet/EfficientNet (Month 3) — for image classification
- YOLO — object detection
- Segment Anything (SAM) — segmentation
- BERT/RoBERTa — text understanding
- Whisper — speech-to-text
- Stable Diffusion — image generation
Your task is to ship these models to production, fine-tune them for your local language (Uzbek), and integrate them with the FastAPI/Django ecosystem.
Start
Start with Introduction to Computer Vision.