Advancing the frontiers of visual understanding through deep learning, medical imaging, autonomous systems, and multimedia analysis
Our computer vision research spans from fundamental algorithms to real-world applications, pushing the boundaries of what machines can see and understand.
Advanced AI systems for medical diagnosis, including retinal imaging, radiology, and pathology analysis.
Computer vision for robotics, autonomous vehicles, and intelligent surveillance systems.
Deep learning for video analysis, content understanding, and multimedia retrieval systems.
Three-dimensional scene understanding, augmented reality, and virtual reality applications.
Our computer vision research leverages the latest advances in deep learning, including transformer architectures, generative models, and self-supervised learning techniques to achieve state-of-the-art performance.
We focus on developing robust, efficient, and interpretable vision systems that can operate in real-world conditions with limited data and computational resources.
Our interdisciplinary approach combines computer vision with domain expertise in healthcare, robotics, and multimedia to create impactful solutions.
Enabling machines to see and understand the world
We employ a comprehensive range of technologies and methodologies to tackle diverse computer vision challenges.
Our computer vision projects span from healthcare applications to autonomous systems, creating real-world impact through advanced visual AI.
Deep learning system for early detection of diabetic retinopathy from retinal images, deployed in rural clinics.
Computer vision algorithms for object detection and scene understanding in autonomous driving systems.
Advanced segmentation algorithms for CT and MRI scans to assist radiologists in diagnosis.
Our computer vision research advances the state-of-the-art in medical imaging, autonomous systems, and multimedia analysis.
We present a comprehensive evaluation of Vision Transformers for medical image segmentation, achieving state-of-the-art performance on multiple medical imaging datasets with 95.2% Dice coefficient.
A robust computer vision system designed specifically for autonomous vehicle perception in complex Indian traffic conditions, handling mixed traffic patterns and challenging weather conditions.
Novel algorithms for real-time 3D scene understanding that enable robust augmented reality applications with accurate object placement and occlusion handling in dynamic environments.
A multi-scale deep learning approach for automated diabetic retinopathy detection that achieves 96.5% sensitivity and 94.2% specificity on a large-scale Indian patient dataset.
Our computer vision research appears in premier conferences and journals
Meet our computer vision research team.
Team component coming soon...