Green leaf




Himadri M - Computer vision developer
Member since:Nov 22, 2020
Profile last updated:Mar 01, 2021
Last activity:March 1, 2021, 9:15 a.m. UTC
Location:
India

Himadri M

Research Engineer

I am an engineer by education and have worn hats of a researcher, a manager, a developer, etc. I have explored the fields of Machine Learning, Deep Learning, Computer Vision during my time in college as well as my experience working in the Industry. I have also touched at some point or other and am comfortable with fields such as Information Retrieval, ML in Security, Augmented Reality, Web Development. Communication among different teams is also something that I have been doing recently.

Skills:   Computer vision  ·  Deep learning  ·  Go  ·  Dart  ·  Robotics  · 
Weekly Availability: 69 hours
Day UTC PST
Sun 03 - 20 19 - 12
Mon 03 - 10 19 - 02
Tues 15 - 22 07 - 14
Wed 15 - 22 07 - 14
Thurs 15 - 22 07 - 14
Fri 15 - 22 07 - 14
Sat 03 - 20 19 - 12

Experience: 3 yrs
Computer vision: 1 yrs
Deep learning: 1 yrs
Engineer's Devices:
Linux
Android
Android Tablet

Experiences

BYJU'S     CV based Line tracing over sheets drawn by kids and Auto-tagging of reference images     Research Engineer     Employment
Oct 2019 - Mar 2020
Education

  • Computer Vision was used to detect lines drawn by kids and filter out if they are drawing gibberish
  • The tagging tool was automated for the reference sheets in Dart
  • Graph theory was used to figure out the strokes in sheets which contains letters, numbers and images in dot tracing format
  • Image processing was used to detect the dots on the canvas from the reference image for auto-tagging
  • Made relevant changes to the back-end in Go to accommodate the requirement

Skills used: Computer vision, Go, Dart


Whodat     Monocular depth estimation and 3-D reconstruction     Deep Learning Engineer     Employment
Oct 2018 - Jan 2019
Technology

  • Deep learning models such as DeMoN and MVDepthNet were used to explore depth using single camera
  • Computer vision was used to stitch together the depths to generate the 3D scene
  • Computer Vision was also used in figuring out the poses between recurring images using visual and dense features

Skills used: Computer vision, Deep learning


UC Berkeley     Neural Programmer Interpreter     Student     Internship
Jun 2017 - Aug 2017
Technology

  • Mujoco simulator was used to simulate a fetch robot and generate data as we didn't have access to the actual robotics system
  • Deep Learning was used to learn from the data generated and generalise to unseen tasks

Skills used: Robotics, Deep learning



Education

IIT(BHU), Varanasi    Integrated Dual Degree (B.Tech and M.Tech)  (Computer Science and Engineering)
Jul 2013 - Jul 2018