The computational intelligence and machine vision laboratory (Vision Lab) aims to develop novel theory, state-of-art algorithms and architectures for learning and real-time applications in the areas of:

Human and Machine-Centric Interaction, Learning and Recognition

Expression Learning

Data-limited domain adaptation and transfer learning for learning latent expression labels of child facial expression images

point-figure

Gender Classification Using Gait Kinematics

deep-recurrent

Generalized object recognition using deep recurrent models

ekg_icon

Analysis and Recognition of 3D Facial Expressions in Pose Variations

asd_icon

Non-intrusive optical imaging of face to probe physiological traits in Autism Spectrum Disorder

ekg_icon

Psychology-Driven Human Movement Analysis for Crowd Sourcing

ekg_icon

Adaptive Local Contrast Enhancement Applied to Super-Resolution Image Recreation

ekg_icon

Robot navigation and object detection/tracking

csrn-ekf

CSRN Maze Traversal with EKF

 

Biomedical Imaging and Signal Analysis

classification architecture

Joint Modeling of RNASEQ and Radiomics Data for Glioma Molecular Characterization and Prediction

classification architecture

Prediction of Molecular Mutations in Diffuse Low-Grade Gliomas using MR Imaging Features

pipeline_icon

Glioma Outcome Prediction And Survival Analysis

ekg_icon

Neurophysiologic Analysis of Human Visual Processing and Interfacing for Automated Robot Navigation

ekg_icon

Improved Brain Tumor Growth Prediction and Segmentation in Longitudinal Brain MRI

ekg_icon

Modeling and Fusion of Atlas-Based Tumor Growth and Feature-Based Models for Normal and Abnormal Brain Tissue Segmentation

ekg_icon

Automatic Segmentation of Brain Tumor and Lesions in MR Images

seizure_icon

Deep Recurrent Neural Network for Seizure Detection

 

Environmental and Geosciences Applications

ekg_icon

Efficient Segmentation of Real-Time Traffic Video for Gas Emissions Estimation

ekg_icon

Sensing and Classification of Aerosols in LiDAR Scattering Plots in the Atmosphere

ekg_icon

Shoreline Detection and Mapping Using DEMs Data and Aerial Photos

 

Prior Projects

ekg_icon

Nonlinear techniques for the enhancement of images captured in extremely dark environments

ekg_icon

Real-time stabilization of shaky video captured under low lighting conditions

ekg_icon

Pediatric fingerprint enhancement comparison and growth projection

ekg_icon

High Performance, Efficient, and Power-Aware Hardware Design for Video Processing Systems

ekg_icon

Real-time “One to Many” Neural Network Classification

ekg_icon

Driver’s assistant for visibility improvement

ekg_icon

Multi-sensor imaging and image fusion

ekg_icon

Network Enabled Feature Search for Person Identification from Databases in Servers Distributed Around the World

ekg_icon

Authentication by facial recognition using low resolution web camera

ekg_icon

Small boat detection for port security applications

ekg_icon

Iris Recognition

ekg_icon

Action Recognition