Machine Learning and Computer Vision | Department of Computer Science and Engineering

Machine Learning and Computer Vision

The primary goal of this research group is to analyze the content of an image or video. We analyze the motion of the moving objects and based on the nature of movements we recognize the event happening in the video. We apply deep learning techniques for motion analysis and prediction. Following are some research problems we work on:

RGB-D Video Content Analysis

  • Sceneflow Estimation
  • Human activity recognition in RGB-D videos

Human Activity and Event Recognition

  • Human activity and event analysis
  • Action and Event recognition in egocentric videos

Image/ Video Quality Assessment and Enhancement

  • Image/ Video Dehazing
  • Assessment of Artistic Quality of Photographs
  • Image Quality Assessment

Video Summarisation

Snehasis Mukherjee