|
I am a fourth-year undergraduate at UC Berkeley studying EECS. I currently work with Professor Avideh Zakhor on deep learning for temporal point clouds. Previously, I briefly worked with Angjoo Kanazawa on 3D human mesh reconstruction. |
|
|
I'm interested in computer vision algorithms that are grounded in the physical scene. Most of my work in this area has been done on 3D vision. |
|
David Deng, Avideh Zakhor 3DV, 2020 pdf / poster / code Predicting future LiDAR frames with a novel class of neural network architectures. |
|
|
|
David Deng*, Howard Zhang* EECS 106A Final Project, Fall 2019 website / video Used a soft tactile sensor to generate 3D point cloud reconstructions of objects. Objects are probed at distinct locations with the sensor, and the point cloud readings are transformed to a global coordinate frame using an AR tag. Implemented using Python, C++, PCL, and ROS. |
|
|
 |
Machine Learning Intern, Qualcomm
Summer 2020 Accelerated neural networks on an audio DSP by implementing hardware and software for low precision matrix operations. Trained neural networks to linearize speaker systems. |
|
Systems Engineering Intern, Northrop Grumman
Summer 2019 Configured and tested network connections for military radio. Wrote Python script and GUI to automate testing process. |
|
|
 |
EECS 16A: Designing Information Devices and Systems I
Teaching Assistant: Fall 2019 Tutor: Spring 2019, Fall 2018 |
|
|