Project Deep Sky

AllSky header

Project Deep Sky explores DNN approaches to generating photorealistic and physically accurate sky models for artistic, scientific and engineering communities.

Key Project Milestones:

  • AllSky: Towards Physically-Based Sky-Modeling For Image Based Lighting
  • Icarus: Full Dynamic Range Sky-Modelling For Image Based Lighting
  • Helios: Coming soon!

Research Focus

Sky models are essential for:

  • Artists: Realistic lighting in 3D graphics and rendering with Image-Based Lighting (IBL)
  • Scientists: Understanding natural light’s impact on biological processes including the human circadian rhythm, plant growth, and animal behaviour.
  • Engineering: A wide range of applications including architectural design, urban planning, and solar energy applications

Our work develops DNN approaches to capture the complexity of real-world skies—including sun positioning, atmospheric conditions, and exposure range—enabling artists, researchers, and engineers to control generation through an intuitive set of parameters and/or modalities.

[Read More]

YGK Modular Ventilator

Code for Life Ventilator Challenge Montreal General Hospital Foundation March – April 2020

  • Participated in a 2-week sprint to design and create a life-saving ventilator, ranking in the top 9 of over 1,000 teams and 2,600 participants.
  • Hosted by Montreal General Hospital Foundation, in collaboration with the Research Institute of the McGill University Health Centre (RI-MUHC), the project was a innovation challenge for a simple, low-cost, easy-to-manufacture and easy-to-maintain ventilator which could be deployed anywhere needed to save lives.
  • Worked with a team of professors, academics, professionals and fellow students, leading design of an intuitive UI interface for system controls and patient safety. Over a period of 8 days, a 7,000 line project was designed, created and tested, consisting of 4,000 lines of python revolving around a tkinter UI. Final submission consisted of 10,000 line project, consisting of 7,500+ LOC python and 900+ LOC Arduino.

Source Code

Tags: COVID