Emeritus ProfessorRamakrishn Mukundan

Analysis of Nuclei Segmentation Methods for Applications in Computational Pathology

Handling Structural Inconsistencies in CT Image Reconstructions

Machine learning classification of histologic grades of breast cancer tissues using whole slide images

Segmentation and Classification Algorithms for Mammographic Image Analysis

Improved vertex skinning algorithms based on dual quaternions

Robust Contour Based Surface Reconstruction for Applications in Medical Imaging

Application of Magnetic Resonance Image Segmentation Algorithms in the Characterization of Parkinson’s Disease

Accurate 3D mesh reconstruction from high resolution CT images

Motion estimation from video using 2D anatomical landmarks

Autonomous mapping in connected 3D spaces

Development and analysis of computer vision solution concepts using game engine based simulations in visually realistic environments

Feature Enhanced Speckle Reduction in Ultrasound Images

Developing a massive real-time crowd simulation framework on the GPU

3D Pose Estimation in Videos Using Convolutional Neural Networks

GPU based algorithms for fluid simulation and rendering

Development of visualization codes for fast rendering of micro-structured geometries for 3D printing

Representation of Artistic Styles using Flock Simulation for Impressionist Non-Photorealistic Rendering

Reconstruction of Overhead Electrical Infrastructure for Dynamic Line Rating using Vehicle-Mounted LIDAR

Comparing terrain rendering algorithms on the GPU

Large Scale Image Processing on Parallel Architectures using OpenCL

Proxemics interaction by recognizing human actions

Image classification using local binary patterns

Similarity and rank learning for large scale specific object recognition

Combining shortest paths, bottleneck paths and matrix multiplication

Applications of dual quaternions in three-dimensional motion representation and interpolation