Our Verticals

AI for Anomalous Events Detection

Visual Object Tracking

Self-supervised or Unsupervised Learning
Vision for Crowd Management
Deep Fake Detection
AI for Graph Data Analysis

AI for Bioinformatics

AI in Healthcare and Medicine
Vision for Fast Image Matching and Search
AI for Remote Sensing
AI for Biometric Identification
AI for Autonomous Driving
Computational Imaging
AI for Multimodal Learning

Visual Human Action Recognition

AI for Synthetic Expression Synthesis
AI for Cloud, Fog, and Cluster Computing
Vision for Human Pose Estimation​
AI for Public Health Emergencies
Moving Objects Segmentation and Background Estimation
AI for Security: Firearms Detection​

• AI in Health and Medicine

• Vision for Anomaly Detection

• Vision for Object Tracking

• AI for Bioinformatics

• Vision for Face and Finger Print Recognition

• Vision for Image Understanding

• Vision for Crowd Management

• Vision for Fake Detection

• Vision for Moving Object Detection

• Vision for Human Pose Detection

• AI for Graph Data Analysis

• Vision for Security

• Vision for Facial Expression Synthesis

• AI for Information Fusion

• AI for Cloud and Fog Computing

• AI for Remote Sensing

AI in Health and Medicine:

• Nucleus Detection and Segmentation

• Nucleus Classification

• Tissue Phenotyping

• Skin Cancer Segmentation

• Prostate Cancer Detection

• ECG Estimation Using Low-Cost Portable PPG Sensors

• Emotion Detection Using ECG Signal

• Semantic Segmentation of Histology Images

• Classification of Gleason Score for Prostate Cancers

• Classification of Benign and Malignant Masses in Mammograms

• Unsupervised WSI Classification

AI for Anomolous Events Detection:

• Weakly Supervised Training to Detect Anomalous Events in Surveillance Videos

• Stabilizing Adversarially Learned One-Class Novelty Detection Using Pseudo Anomalies

• Unsupervised Video Anomaly Detection

• An anomaly detection system via moving surveillance robots with human collaboration

• Cleaning Label Noise with Clusters for Minimally Supervised Anomaly Detection

AI for Visual Object Tracking:

• Information Fusion for Visual Object Tracking

• Video Object Segmentation

• Graph Regularized Discriminative Correlation Filter for Visual Object Tracking

• Robust Tracking via Feature Enrichment

• Robust Structural Low-Rank Tracking

• Improving Object Tracking by Added Noise and Channel Attention

• Feature fusion with channel and spatial attention for robust visual object tracking

• Adaptive Feature Selection Siamese Networks for Visual Tracking

• Complete Moving Object Detection in the Context of Robust Subspace Learning

AI for Human Pose Detection:

• Quantification of Occlusion Handling Capability of 3D Human Pose Estimation Framework

• A Boosting Framework for Human Posture Recognition

• Walk Like Me: Video to Video Action Transfer

• Human fall detection

AI for Graph Data Analysis:

• Higher-Order Sparse Convolutions in Graph Neural Networks

• Reconstruction of Time-Varying Graph Signals

• Network Community Detection

AI for Expression Synthesis:

• Improving GAN Generalization via Regressed Conditioning for Facial Expression Synthesis

• Masked Linear Regression for Learning Local Receptive Fields for Facial Expression Synthesis

AI for Cloud, Fog and Cluster Computing:

• Predictive Autoscaling of Microservices Hosted in Fog Microdata Center

• Predictive Auto-scaling of Multi-tier Applications Using Performance Varying Cloud Resources

• Canny edge detection and Hough transform for high resolution video streams using Hadoop and Spark

• Web application resource requirements estimation based on the workload latent features

• Dynamic workload patterns prediction for proactive auto-scaling of web applications

AI for Remote Sensing:

• Improving Chlorophyll-a Estimation from sentinel-2 (MSI) in the Barents Sea using Machine Learning

• Ocean Color Net (OCN) for the Barents Sea

AI for Autonomous Driving:

• Statistically correlated multi-task learning for autonomous driving

AI for Speech and Text Processing:

• Single branch Network for Multimodal Training

• Cross-modal Speaker Verification and Recognition: A Multilingual Perspective

• Do Cross Modal Systems Leverage Semantic Relationships?

• Deep Latent Space Learning for Cross-modal Mapping of Audio and Visual Signals

AI for Security: Firearms Detection:

• Detection and Localization of Firearm Carriers in Complex Scenes for Improved Safety Measures

• Leveraging orientation for weakly supervised object detection with application to firearm localization

• Localizing firearm carriers by identifying human-object pairs

•  Orientation Aware Object Detection with Application to Firearms

• Improving security surveillance by hidden cameras

AI for Bioinformatics:

• EnProCell: Ensembled lower-dimensional projections of cellular expression improves the cell type classification from single-cell RNA sequencing

• Deep-Ace: LSTM-based Prokaryotic Lysine Acetylation Site Predictor

• Data science meets high-throughput single-cell sequencing, in particular, single-cell RNA sequencing and spatial transcriptomics