Previously, I completed my Bachelor's degree in Computer Science at Jacobs University Bremen. During my Bachelor's, I spent a term abroad at Carnegie Mellon University, where I did independent research under the supervision of Prof. Ruslan Salakhutdinov.
I'm interested in working at the intersection of 3D computer vision, computer graphics, and machine learning. My Master's thesis focused on the 3D GAN inversion for 3D reconstruction and novel view synthesis from a single 2D image.
EG3D is a powerful {z, camera}->image generative model, but inverting EG3D (finding a corresponding z for a given image) is not always trivial. We propose a fully-convolutional encoder for EG3D based on the observation that predicting both z code and tri-planes is beneficial. TriPlaneNet also works for videos and in real time (check out the Live Demo).
The framework leverages a conditional Generative Adversarial Network (cGAN) to manipulate the style of a specific category within a given image and compares segmentation maps to verify the model's robustness.
By using neural networks as encoding functions, the method automatically learns the quantum-level interaction logic of the system from the data to compute and infer macro-level properties of chemical or material systems.
The webpage template was borrowed from the exciting page of Jon Barron.