NeRF: Neural Radiance Fields
A complete NeRF pipeline for novel view synthesis, representing 3D scenes as continuous volumetric functions learned from multi-view images.
UC Berkeley EECS graduate with professional and research experience in deep learning, computer vision, and signal processing. I have a passion for building AI systems that solve real-world problems and give us a better understanding of the world around us.
A complete NeRF pipeline for novel view synthesis, representing 3D scenes as continuous volumetric functions learned from multi-view images.
Exploring generative image models through flow matching on MNIST and creative applications like visual anagrams, hybrid images, and inpainting.
A robotic system using a Sawyer arm to autonomously play pool, combining computer vision for ball detection with motion planning for shot execution.
A mechanistic interpretability study investigating how large language models represent and pursue instrumental goals through linear probing and activation analysis.
An instructional operating system implementing multithreading, synchronization, user programs with system calls, and a file system supporting concurrency.
Creating seamless panoramic mosaics through homography estimation, with both manual correspondence and automatic feature matching using Harris corners and RANSAC.
Exploring filters and frequencies for edge detection, image sharpening, hybrid images, and multiresolution blending using Gaussian and Laplacian stacks.
Reconstructing color photographs from Prokudin-Gorskii's historic glass plate negatives by aligning RGB channels using image pyramids and NCC metrics.
Open to all opportunities in AI Engineering, Software Engineering, and MLOps. Feel free to reach out.
zpricz@berkeley.edu →`