ProjectsMarch 1, 2024

Human Emotion Classifier

Overview

Human Emotion Classifier is a computer vision application that detects human emotions from facial images in real time. Users upload or capture a photo through a web interface, and the system classifies the expression into categories such as happy, sad, angry, surprised, and neutral. The project combines deep learning model training with a practical full-stack deployment, making the classifier accessible through a browser without requiring any local ML setup.

Tech Stack

The model is a convolutional neural network (CNN) trained with TensorFlow and Keras on labeled facial expression datasets. A Flask API serves the trained model, handling image preprocessing, inference, and JSON responses. The React frontend provides image upload, live preview, and results display with clear confidence indicators for each emotion class.

Highlights

Trained a CNN architecture optimized for facial emotion recognition across multiple expression classes.
Built a Flask inference API with image preprocessing pipeline for consistent model input.
Created a React frontend for intuitive image upload and real-time classification results.
End-to-end pipeline from raw image input to labeled emotion output in a single user flow.

More projects

Python

Deep Researcher RAG System

Multi-agent research pipeline with FAISS vector store, Streamlit UI, and FastAPI backend. Generates comprehensive research reports from a single query using RAG architecture.

Case study View project

Next.js

IdeaClinic Forum

Full-featured forum platform built for DEI, SRMIST using Next.js and Supabase for real-time data and authentication.

Case study View project

TensorFlow

Video Action Recognition

CNN-based action recognition model using MovieNet architecture for Samsung Prism. Achieved 97% training accuracy on compressed video datasets.

Case study