The rapid advancement in self-supervised representation learning has highlighted its potential to leverage unlabeled data for learning rich visual representations. However, the existing techniques, particularly those employing different augmentations of the same image, often rely on a limited set of simple transformations that cannot fully capture variations in the real world. This constrains the diversity and quality of samples, which leads to sub-optimal representations. In this paper, we introduce a framework that enriches the self-supervised learning (SSL) paradigm by utilizing generative models to produce semantically consistent image augmentations. By directly conditioning generative models on a source image, our method enables the generation of diverse augmentations while maintaining the semantics of the source image, thus offering a richer set of data for SSL. Our extensive experimental results on various joint-embedding SSL techniques demonstrate that our framework significantly enhances the quality of learned visual representations by up to 10% Top-1 accuracy in downstream tasks. This research demonstrates that incorporating generative models into the joint-embedding SSL workflow opens new avenues for exploring the potential of synthetic data. This development paves the way for more robust and versatile representation learning techniques.

Novelty detection is the task of recognizing samples that do not belong to the distribution of the target class. During training, the novelty class is absent, preventing the use of traditional classification approaches. Deep autoencoders have been widely used as a base of many novelty detection methods. In particular, context autoencoders have been successful in the novelty detection task because of the more effective representations they learn by reconstructing original images from randomly masked images. However, a significant drawback of context autoencoders is that random masking fails to consistently cover important structures of the input image, leading to suboptimal representations - especially for the novelty detection task. In this paper, to optimize input masking, we introduce a Mask Module that learns to generate optimal masks and a Reconstructor that aims to reconstruct masked images. The networks are trained in an adversarial setting in which the Mask Module seeks to maximize the reconstruction error that the Reconstructor is minimizing. When applied to novelty detection, the proposed approach learns semantically richer representations compared to context autoencoders and enhances novelty detection at test time through more optimal masking. Novelty detection experiments on the MNIST and CIFAR-10 image datasets demonstrate the proposed approach’s superiority over cutting-edge methods. In a further experiment on the UCSD video dataset for novelty detection, the proposed approach achieves a frame-level Area Under the Curve (AUC) of 99.02% and an Equal Error Rate (EER) of 5.4%, exceeding recent state-of-the-art models. Code available at https://github.com/jewelltaylor/OLED.

Generative adversarial networks (GANs) synthesize realistic images from random latent vectors. While many studies have explored various training configurations and architectures for GANs, the problem of inverting the generator of GANs has been inadequately investigated. We train a ResNet architecture to map given faces to latent vectors that can be used to generate faces nearly identical to the target. We use a perceptual loss to embed face details in the recovered latent vector while maintaining visual quality using a pixel loss. The vast majority of studies on latent vector recovery are very slow and perform well only on generated images. We argue that our method can be used to determine a fast mapping between real human faces and latent-space vectors that contain most of the important face style details. At last, we demonstrate the performance of our approach on both real and generated faces.