Computer Vision β’ Generative AI β’ Trustworthy AI
I'm a B.Tech graduate in Artificial Intelligence & Data Science passionate about building intelligent systems that solve real-world problems.
My interests include:
- π§ Deep Learning & Computer Vision
- π AI-Generated Media Detection
- π€ Generative AI Applications
- π Data Science & Machine Learning
- π Research and Technical Writing
I enjoy transforming research concepts into practical, deployable AI solutions while emphasizing interpretability, reliability, and ethical AI.
- Research in AI-generated image detection and image forensics
- Building trustworthy and explainable AI systems
- Exploring Computer Vision and Generative AI
- Developing end-to-end AI applications
- Contributing to open-source AI projects
- π B.Tech Graduate in Artificial Intelligence & Data Science
- π₯ Second Place β Department-Level main project Competition
- π Research-oriented projects in AI Image Forensics
- π Developed DeepReveal for pixel-level manipulation localization
- π» Built AI-powered web applications for real-time inference
- π§ Experience in Deep Learning, Computer Vision, and Data Science
| Project | Description |
|---|---|
| π DeepReveal | Pixel-level AI image manipulation detection and localization using U-Net segmentation |
| π°οΈ Cross-Seasonal Image Matching | Robust matching of satellite and drone imagery across seasonal variations using SuperPoint feature extraction and geometric verification |
| πΌοΈ Deepfake Detection Web Application | MobileNetV3-Large based system for detecting GAN-generated images with a Streamlit interface |
A deep learning-based image forensics framework that detects and localizes AI-generated or manipulated regions within images using semantic segmentation. Unlike traditional real/fake classifiers, DeepReveal highlights exactly where modifications occur, improving transparency and interpretability.
Tech Stack: Python β’ PyTorch β’ U-Net β’ OpenCV β’ Flask
Applications: Digital Forensics β’ Media Authentication β’ Deepfake Localization β’ Content Verification
A computer vision pipeline for matching satellite and drone imagery across different seasons using SuperPoint feature extraction and geometric verification, enabling robust correspondence despite environmental and viewpoint changes.
Tech Stack: Python β’ OpenCV β’ PyTorch β’ SuperPoint
Applications: Remote Sensing β’ Geospatial AI β’ Visual Localization
A deep learning-based image classification system for detecting GAN-generated face images.
Tech Stack: Python β’ PyTorch β’ MobileNetV3-Large β’ Streamlit β’ OpenCV
Applications: AI-Generated Image Detection β’ Content Verification β’ Digital Media Analysis
- π Trustworthy & Explainable AI
- π High-Quality Data & Preprocessing
- π Reproducible Machine Learning Experiments
- π Research & Technical Writing
- π Ethical and Responsible AI Development
β If you find my work useful, consider starring a repository β it helps support my projects and research!