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
Variational Autoencoder for Data Generation
Learns a smooth latent space of images and samples new ones from it
2-3 weeks build
Overview
A convolutional VAE is trained on MNIST and Fashion-MNIST, with the reparameterisation trick and KL term implemented and explained step by step. The kit compares it with a plain autoencoder, visualises the latent space with t-SNE, and provides sliders to walk through latent dimensions and watch generated images change.
What makes it stand up
- 01Convolutional encoder and decoder with reparameterisation
- 02Plain autoencoder baseline to show what the KL term adds
- 03Beta-VAE variant for more disentangled factors
- 04t-SNE plots of the learned latent space
- 05Interactive latent sliders for generation
Module breakdown
- 01
Data
MNIST and Fashion-MNIST loaders.
- 02
Autoencoder
Deterministic baseline with reconstruction loss.
- 03
VAE
Probabilistic encoder, sampling layer and ELBO loss.
- 04
Latent analysis
t-SNE projections and interpolation grids.
- 05
Explorer
Streamlit sliders over latent dimensions.
After this, you will be able to
- Explain the reparameterisation trick and why it is needed
- Compare a VAE with a GAN for generation
- Demo moving through latent space live
Technology stack
- Python
- PyTorch
- NumPy
- scikit-learn
- Streamlit
- Level
- Major Project
- Domain
- Artificial Intelligence
- Build time
- 2-3 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