LARGE: Latent-Based Regression through GAN Semantics

07/22/2021
by   Yotam Nitzan, et al.
6

We propose a novel method for solving regression tasks using few-shot or weak supervision. At the core of our method is the fundamental observation that GANs are incredibly successful at encoding semantic information within their latent space, even in a completely unsupervised setting. For modern generative frameworks, this semantic encoding manifests as smooth, linear directions which affect image attributes in a disentangled manner. These directions have been widely used in GAN-based image editing. We show that such directions are not only linear, but that the magnitude of change induced on the respective attribute is approximately linear with respect to the distance traveled along them. By leveraging this observation, our method turns a pre-trained GAN into a regression model, using as few as two labeled samples. This enables solving regression tasks on datasets and attributes which are difficult to produce quality supervision for. Additionally, we show that the same latent-distances can be used to sort collections of images by the strength of given attributes, even in the absence of explicit supervision. Extensive experimental evaluations demonstrate that our method can be applied across a wide range of domains, leverage multiple latent direction discovery frameworks, and achieve state-of-the-art results in few-shot and low-supervision settings, even when compared to methods designed to tackle a single task.

READ FULL TEXT

page 19

page 20

page 21

page 25

page 26

page 27

page 28

page 30

research
05/26/2021

Disentangled Face Attribute Editing via Instance-Aware Latent Space Search

Recent works have shown that a rich set of semantic directions exist in ...
research
11/16/2021

Delta-GAN-Encoder: Encoding Semantic Changes for Explicit Image Editing, using Few Synthetic Samples

Understating and controlling generative models' latent space is a comple...
research
12/16/2021

Self-supervised Enhancement of Latent Discovery in GANs

Several methods for discovering interpretable directions in the latent s...
research
11/29/2021

Latent Transformations via NeuralODEs for GAN-based Image Editing

Recent advances in high-fidelity semantic image editing heavily rely on ...
research
10/28/2021

Multi-Attribute Balanced Sampling for Disentangled GAN Controls

Various controls over the generated data can be extracted from the laten...
research
07/16/2023

Householder Projector for Unsupervised Latent Semantics Discovery

Generative Adversarial Networks (GANs), especially the recent style-base...
research
05/23/2023

Exploring Semantic Variations in GAN Latent Spaces via Matrix Factorization

Controlled data generation with GANs is desirable but challenging due to...

Please sign up or login with your details

Forgot password? Click here to reset