Search Results for author:
Found 4 papers, 2 papers with code
SASSL: Enhancing Self-Supervised Learning via Neural Style Transfer
Existing data augmentation in self-supervised learning, while diverse, fails to preserve the inherent structure of natural images.
Making Vision Transformers Truly Shift-Equivariant
For computer vision, Vision Transformers (ViTs) have become one of the go-to deep net architectures.
Learnable Polyphase Sampling for Shift Invariant and Equivariant Convolutional Networks
We propose learnable polyphase sampling (LPS), a pair of learnable down/upsampling layers that enable truly shift-invariant and equivariant convolutional networks.
Inverting Adversarially Robust Networks for Image Synthesis
Despite unconditional feature inversion being the foundation of many image synthesis applications, training an inverter demands a high computational budget, large decoding capacity and imposing conditions such as autoregressive priors.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
