Gaurav Kuppa
COMPUTER VISION · APPLIED MACHINE LEARNING · ROBOTICS · CLIMATE CHANGE
(408) 963 - 7074 adiu5l@r.postjobfree.com gauravkuppa gauravkuppa Education
San Jose State University August 2017 - December 2020 B.S. IN COMPUTER ENGINEERING GPA: 3.7/4.0
• Student Organizations: ML@SJSU, Software and Computer Engineering Society, SJSU Robotics, Theta Tau
• Relevant Coursework: Computer Architecture & Organization, Embedded Systems, Data Structures (Python and C++) Experience
Machine Learning Intern June. 2020 - Sept. 2020
AEROSPACE CORPORATION Los Angeles, CA, USA
• Testeddifferentreinforcementlearningalgorithms,Clipped PPO, DQNandDDQN,in robotic applications by using RL-Coach, ROS and Gazebo to simulate space satellite rendezvous
• Streamlined communication between AWS RoboMaker and AWS SageMaker and ran Bayesian hyperparameter tuning to im- prove training
• Implemented time-series alignment algorithm; demonstrated a 84% decrease in dynamic time warping distance Generative Modeling Research Assistant January 2020 - Present SAN JOSE STATE UNIVERSITY San Jose, CA, USA
• Worked with Dr. Ziwei Liu and Dr. Teng Moh to design virtual try-on network with self-attention, DensePose annotations, and smooth activation functions to improve quality of garment transfer while reducing training time and memory usage
• Conducted research about generative adversarial networks, image-to-image translation, 3D modeling, and style transfer
• Investigated novel ideas using PyTorch to achieve precise control of generated media; our methods created a 12% increase in try-on quality with greater body and cloth texture detail Software Engineering Research Assistant June 2018 - June 2020 SAN JOSE STATE UNIVERSITY San Jose, CA, USA
• Worked with Dr. Mohamed Fayad on ”Unified Software Architecture for Stable Machine Learning” by determining functional and nonfunctional requirements and designing stable pattern language - which resulted in a Software Stability Model
• Executed scenario-based testing to define core knowledge through functional and non-functional requirements for unified soft- ware architecture
• Designed Unified Modeling Languages(UML) and Software Stability Models using MS Visio Projects
Autonomous Drone (Link) January 2020 - Present
SAN JOSE STATE UNIVERSITY San Jose, CA, USA
• Created high-level software architecture to communicate between SJ2 microcontroller and Raspberry Pi 4
• Implemented real-time embedded flight controller; utilized cascade control loop and quaternion rotations with FreeRTOS and C
• Demonstrate simulated general-purpose prototype of a delivery drone using ROS and Gazebo Autonomy Lead August 2019 - August 2020
SAN JOSE STATE UNIVERSITY ROBOTICS San Jose, CA, USA
• Used OpenCV to implement image pre-processing and used TensorFlow to implement YoloV3 with DarkNet backbone
• Trained real-time pole detection model using 16-bit precision training and deployed onto NVIDIA Jetson Nano using ROS to in- form robot’s actions using visual insights
Publications
2021
Kuppa. G, Jong, A., Liu, V., Liu, Z., and Moh, T., ShineOn: Illuminating Design Choices for Practical Video-based Virtual Clothing Try-on. To Appear at Generation of Human Behavior Workshop at WACV 2021 2020 Fayad, M.E., and Kuppa, G, Stable Machine Learning Knowledge Map Domain Analysis FTC 2020 Vancouver, Canada 2019 Fayad, M.E., Kuppa, G, and Hamu, D., Unified and Stable Privacy Model. iCiCSE2019 Miri, Malaysia 2019 Fayad, M.E., Kuppa, G, and Hamu, D., Unified and Stable Privacy Model. IJATCSE Vol.8, No.4, Article 80 2019 Fayad, M.E., Kuppa, G, and Hamu, D., Unified and Stable Project – “Ushering in the Future”. FTC 2019 San Francisco, USA 2019 Fayad, M.E., Kuppa, G, Jindal, S., and Hamu, D., A Cutting-Edge Unified and Stable Rule Design Pattern. FTC 2019 San Francisco, USA Awards
2020 1st Place in Video Virtual Try-On Challenge, 2020 Conference on Computer Vision and Pattern Recognition 2019 Dean Scholar, San Jose State University
2018 SVIC Showcase Finalist, San Jose State University