SETGAN: Scale and Energy Trade-off GANs for Image Applications on Mobile Platforms

by   Nitthilan Kannappan Jayakodi, et al.

We consider the task of photo-realistic unconditional image generation (generate high quality, diverse samples that carry the same visual content as the image) on mobile platforms using Generative Adversarial Networks (GANs). In this paper, we propose a novel approach to trade-off image generation accuracy of a GAN for the energy consumed (compute) at run-time called Scale-Energy Tradeoff GAN (SETGAN). GANs usually take a long time to train and consume a huge memory hence making it difficult to run on edge devices. The key idea behind SETGAN for an image generation task is for a given input image, we train a GAN on a remote server and use the trained model on edge devices. We use SinGAN, a single image unconditional generative model, that contains a pyramid of fully convolutional GANs, each responsible for learning the patch distribution at a different scale of the image. During the training process, we determine the optimal number of scales for a given input image and the energy constraint from the target edge device. Results show that with SETGAN's unique client-server-based architecture, we were able to achieve a 56 for a loss of 3 training, we obtain around 4x gain in training time on the server.



There are no comments yet.


page 3

page 4

page 6

page 7

page 8


SinGAN: Learning a Generative Model from a Single Natural Image

We introduce SinGAN, an unconditional generative model that can be learn...

TinyGAN: Distilling BigGAN for Conditional Image Generation

Generative Adversarial Networks (GANs) have become a powerful approach f...

Improved Training of Generative Adversarial Networks Using Representative Features

Despite of the success of Generative Adversarial Networks (GANs) for ima...

Tabular GANs for uneven distribution

GANs are well known for success in the realistic image generation. Howev...

Kernel Mean Matching for Content Addressability of GANs

We propose a novel procedure which adds "content-addressability" to any ...

Diverse Generation from a Single Video Made Possible

Most advanced video generation and manipulation methods train on a large...

Hijack-GAN: Unintended-Use of Pretrained, Black-Box GANs

While Generative Adversarial Networks (GANs) show increasing performance...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.