PopularMajor Project
Artificial Intelligence
Face Recognition Attendance System
Camera-based attendance with liveness detection and analytics
- Python
- OpenCV
- FaceNet
- Flask
- +2
₹4,900₹5,999
4.8/301 reviews
Artificial Intelligence · Major Project
Image Segmentation with U-Net
Pixel-level segmentation, classical methods versus a trained U-Net
3-4 weeks build
Overview
Starts with classical segmentation — Otsu thresholding, k-means colour clustering and watershed — then trains a U-Net from scratch on the Oxford-IIIT Pet dataset's pixel masks. Every method is scored on the same test images with IoU and Dice, so the report shows exactly where learned features beat hand-tuned rules and where they do not.
What makes it stand up
- 01Three classical baselines implemented and tuned by hand
- 02U-Net with skip connections written layer by layer in PyTorch
- 03Combined Dice and cross-entropy loss for class imbalance
- 04IoU and Dice scored per image with failure-case galleries
- 05Interactive page that overlays predicted masks on uploads
Module breakdown
- 01
Classical baselines
Otsu, k-means and marker-based watershed in OpenCV.
- 02
Dataset pipeline
Paired image and mask augmentation that stays aligned.
- 03
U-Net model
Encoder-decoder with skip connections and batch normalisation.
- 04
Training
Mixed loss, checkpointing and validation IoU tracking.
- 05
Comparison
Metric tables and side-by-side mask overlays for every method.
After this, you will be able to
- Explain why skip connections recover fine boundaries
- Defend IoU and Dice as the right metrics for segmentation
- Demo live mask overlays against the classical baselines
Technology stack
- Python
- PyTorch
- OpenCV
- scikit-image
- Streamlit
- Level
- Major Project
- Domain
- Artificial Intelligence
- Build time
- 3-4 weeks
- Difficulty
You receive
- Complete, commented source code
- 60-90 page project report (IEEE format)
- Editable presentation deck
- Architecture, ER and UML diagrams
- Local setup and deployment guide
- Sample dataset or seed data
- Viva question bank with answers
- One 45-minute walkthrough call
- 30 days of doubt-clearing support