Published on Mon Nov 02 2020

Learning a Deep Reinforcement Learning Policy Over the Latent Space of a Pre-trained GAN for Semantic Age Manipulation

Kumar Shubham, Gopalakrishnan Venkatesh, Reijul Sachdev, Akshi, Dinesh Babu Jayagopi, G. Srinivasaraghavan

Generative Adversarial Networks (GANs) have shown promising results in generating high fidelity images. However, studies to understand the semantic layout of the latent space of pre-trained models are still limited.

0
0
0
Abstract

Learning a disentangled representation of the latent space has become one of the most fundamental problems studied in computer vision. Recently, many Generative Adversarial Networks (GANs) have shown promising results in generating high fidelity images. However, studies to understand the semantic layout of the latent space of pre-trained models are still limited. Several works train conditional GANs to generate faces with required semantic attributes. Unfortunately, in these attempts, the generated output is often not as photo-realistic as the unconditional state-of-the-art models. Besides, they also require large computational resources and specific datasets to generate high fidelity images. In our work, we have formulated a Markov Decision Process (MDP) over the latent space of a pre-trained GAN model to learn a conditional policy for semantic manipulation along specific attributes under defined identity bounds. Further, we have defined a semantic age manipulation scheme using a locally linear approximation over the latent space. Results show that our learned policy samples high fidelity images with required age alterations, while preserving the identity of the person.

Sat May 15 2021
Computer Vision
AgeFlow: Conditional Age Progression and Regression with Normalizing Flows
Age progression and regression aim to synthesize photorealistic appearance of a given face image with aging and rejuvenation effects, respectively. Existing Generative adversarial networks (GANs) based methods suffer from the following major issues: unstable training introducing strong ghost artifacts in generated faces.
1
0
0
Wed Mar 06 2019
Computer Vision
Age Progression and Regression with Spatial Attention Modules
Age progression and regression refers to aesthetically render-ing a given face image to present effects of face aging and rejuvenation. We propose a framework based on conditional Generative Adversarial Networks (cGANs) to achieve age progress and regression simultaneously.
0
0
0
Tue Feb 07 2017
Computer Vision
Face Aging With Conditional Generative Adversarial Networks
Generative Adversarial Networks (GANs) can produce synthetic images of exceptional visual fidelity. We make a particular emphasize on preserving the original person's identity in the aged version of his/her face.
0
0
0
Fri Feb 26 2021
Computer Vision
Continuous Face Aging Generative Adversarial Networks
0
0
0
Thu Feb 01 2018
Computer Vision
Face Aging with Contextual Generative Adversarial Nets
Face aging, which renders aging faces for an input face, has attracted extensive attention in the multimedia research. Several conditional Generative Adversarial Nets (GANs) based methods have achieved great success. However, these methods fail to capture the gradual shape and texture changes between adjacent age groups.
0
0
0
Wed Sep 11 2019
Computer Vision
How Old Are You? Face Age Translation with Identity Preservation Using GANs
We present a novel framework to generate images of different age while preserving identity information, which is known as face aging. Different from recent popular face aging networks, our approach do not simply transfer a young face to an old one.
0
0
0