About Me
A computer whisperer who loves communicating with people as much as machines. My research is focused towards deep learning for computer vision and I have a keen interest in application of everyday research to real world scenarios. My prior experience is spread across various problem domains such as face recognition, object recognition and detection, metric learning, scaling and optimizing pre-existing code. Apart from research I love designing user facing products and share an enthusiasm for managing painless project deliveries, if such a world exists :).
You do not need to know everything, everything you need to know you will figure it out when you need to know it.
Publications Complete list @
- Reducing Network Agnostophobia
Neural Information Processing Systems (NeurIPS) 2018 - Spotlight Presentation Akshay Raj Dhamija, Manuel Günther, Terrance E. Boult
- The overlooked elephant of object detection: Open set
Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. 2020 Akshay Raj Dhamija, Manuel Günther, Jonathan Ventura, Terrance E. Boult
- Improving Deep Network Robustness to Unknown Inputs with Objectosphere
CVPR Workshops, Uncertainty and Robustness in Deep Visual Learning 2019 Akshay Raj Dhamija, Manuel Günther, Terrance E. Boult
- Unconstrained Face Detection & OpenSet Recognition Challenge
IEEE International Joint Conference on Biometrics (IJCB), 2017 M Günther, P Hu, C Herrmann, CH Chan, M Jiang, S Yang, AR Dhamija, D Ramanan, J Beyerer, J Kittler, MA Jazaery, MI Nouyed, G Guo, C Stankiewicz, TE Boult
- Watchlist Adaptation: Protecting the Innocent
International Conference of the Biometrics Special Interest Group (BIOSIG), 2020 Manuel Günther, Akshay Raj Dhamija, Terrance E Boult
- What's Hiding in my Deep Features?
Book Chapter, Deep Learning in Biometrics, 2018 Ethan M Rudd, Manuel Günther, Akshay R Dhamija , Faris A Kateb, Terrance E Boult
Work Experience
- Research Assistant Vision And Security Technology (VAST) Lab, September 2015 - Present
- Computer Vision Intern Neon, STAR Labs (Samsung Research America) California, December 2019 - Feb 2020
- Computer Vision Intern Misty Robotics , June 2018 - August 2018
- Project Consultant My Personal Health Records Express (MphRx), November 2012 - August 2015
Project Portfolio
Feature Server
A flask server used to amortize feature extraction cost to help run ablation studies using features from any pre-trained network.
Various Algorithms & Software Tools (VAST)
Initiated by me to address our labs troubles with reproducibility, redudandant code, continued support and collaboration. Currently used in majority of our projects and those of our collaborators.
GRE : App
Born out of my troubles with reading comprehension and designed to help users study on the go. Last checked 5K+ installs spread across 20+ countries.
VR Website with AFrame
Curiosity to develop VR content and learning A-Frame
Education
A few accomplishments
Invited on UCCS panel by CU board of Trustees
Presenting and winning top award at Mountain Lion Research day