Selected Publications
Diffusion Models for Outfit Rendering: Novel Conditioning Architectures for Subject-driven Generation
Vignesh Srinivasan, Nikolay Jetchev, Martin Heusel, Tofigh Naghibi
ECCV Workshops, 2022
We propose novel conditioning architectures for diffusion models for generating curated outfits to be rendered on a digital human in predefined pose.
Langevin Cooling for Unsupervised Domain Translation
Vignesh Srinivasan, Klaus-Robert Müller, Wojciech Samek, Shinichi Nakajima
IEEE Transactions on Neural Networks and Learning Systems, 2022
We propose to perform Langevin dynamics on fringe samples to lower the temperature of test samples before applying the base domain translation method resulting in a significantly improved performance.
Robustifying Models Against Adversarial Attacks by Langevin Dynamics
Vignesh Srinivasan, Csaba Rohrer, Arturo Marban, Klaus-Robert Müller, Wojciech Samek, Shinichi Nakajima
Neural Networks, 2021
We introduce a generative model of the conditional distribution of the inputs given labels that can be learned through a supervised Denoising Autoencoder (sDAE) in alignment with a discriminative classifier.