Research Areas


My research is focused on computer vision, machine learning, deep learning, signal processing and optimization. Specifically, I worked on:

  • Image and video analysis: Primary used denoising, hough transforms, super resolution etc.

  • Object detection:

  1. Used state of the art deep learning single shot detector and you only look once models to detect people in surveillance cameras

  2. Fine tuned the YOLO, SSD for high performance object detection in images with heavily overlapping instances

  • Multiple object tracking:

  1. Exploited higher-order information such as long-term motion and appearance models for tracking using multi hypothesis theorem in a tracking-by-detection framework

  2. Experimented with SORT and Deep SORT pragmatic approach to multiple object tracking with a focus on simple, effective algorithms

  • Human pose estimation to detect peculiar key points on human body parts and further, used it for locating multiple regions of interest

  • Image classification

  • Image clustering using features detectors and descriptors such as SIFT, SURF, HOG, bag of words approaches and unsupervised deep representational learning such as Autoencoders

  • Predictive modelling: Development of forecasting models using various data analytics techniques for prediction of COVID-19 time series

  • Risk staging and survival predictions in Multiple Myeloma blood cancer: development of classification models for categorization of cancer patients into multiple risk groups (primarily used birch clustering, Gaussian mixture model clustering, aggolomerative clustering, decision trees, random forest, xgboost, gradient boosted trees, support vector machines, elastic net etc. on operational systems.

  • Modeled and implemented novel algorithms in regression modelling for brain region's connectivity analysis with multiple constrained regularizations and reconstruction

Currently, I am actively working towards anamoly detection, face recognition, video summarization, under vehicles images clustering, deep learning models for sleep apnea detection using ECG data, COVID-19 risk severity analysis and interpretation