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
Human Face Generation with DCGAN
Trains a DCGAN on CelebA to generate new, non-existent faces
3-4 weeks build
Overview
A deep convolutional GAN is built from scratch and trained on the CelebA face dataset at 64x64 resolution. The kit documents the training dynamics — generator and discriminator losses, mode collapse and the stabilising tricks that fix it — and scores sample quality with Fréchet Inception Distance at checkpoints across training.
What makes it stand up
- 01Generator and discriminator written layer by layer
- 02Training progress grid saved every epoch
- 03Label smoothing and other stabilisation tricks explained
- 04Latent-space interpolation between generated faces
- 05FID tracked across checkpoints in the report
Module breakdown
- 01
Data
CelebA download, centre crop and normalisation.
- 02
Generator
Transposed convolutions with batch normalisation.
- 03
Discriminator
Strided convolutions with LeakyReLU.
- 04
Training
Adversarial loop with checkpoints and loss logging.
- 05
Explorer
Streamlit page for sampling and interpolation.
After this, you will be able to
- Explain the minimax game between generator and discriminator
- Diagnose mode collapse from training curves
- Demo generating and morphing faces live
Technology stack
- Python
- PyTorch
- torchvision
- NumPy
- 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