📅 This bar shows how far I’ve come since Day 1
Every day counts. The tracker below automatically updates based on the number of days since I started this journey on July 8, 2025.
Week 1 – CV Fundamentals
- well today was day one i watched standford university lectures and started with the fundamentals by the help of campusx and the book on programming computer vision by Jan Erik Solem.
- today is was the second day and i have done comparision between cnn and ann , i have also learned to perform back propagation on cnn.
- made a minor project around image classification , learned data augumentation and worked with pretrained models around ImageNET
Week 2 – Transfer Learning & Evaluation
- Use pretrained ResNet/VGG
- Understand accuracy, precision, recall
- Visualize predictions
Week 3 – Object Detection
- YOLOv5, SSD architecture
- Train YOLOv5 on custom data
- Draw bounding boxes
Week 4 – Image Segmentation
- Learn FCN, U-Net, DeepLabV3
- Binary segmentation using U-Net
- Train model on medical dataset
Week 5 – Generative Models (GANs)
- Study DCGAN, Pix2Pix, CycleGAN
- Implement DCGAN on CelebA
- Understand adversarial loss
Week 6–7 – DeepFill v2 Paper
- Read DeepFill v2 paper
- Clone repo, run test samples
- Train on inpainting dataset
- Test real-world images
Week 8 – Research Experiments
- Modify DeepFill architecture
- Apply to new domain (sketch/satellite)
- Write blog or paper
Week 9+ – Vision Transformers
- ViT, Swin Transformer
- Learn SimCLR, BYOL
- Multimodal project (OCR + LLM)
- Begin arXiv/CVPR paper prep