Questions tagged [generative-adversarial-networks]
For questions related to generative adversarial networks (GANs), introduced in the paper Generative Adversarial Nets (2014) by J. Goodfellow et al. A GAN is composed of a discriminative model (D) and a generative model (G). The discriminator D needs to distinguish between data generated by the generator G and data in the training set, while the generator G needs to generate data such that the discriminator D is not able to accomplish its task.
257
questions
1
vote
0
answers
24
views
cGAN: Discriminator loss going to zero while Generator's going always up but the result is very good
I have a Conditional Generative Adversarial Network for Quantum State Tomography. The metrics I am monitoring during the training process are the losses and the Fidelity (the degree of similarity ...
- 11
1
vote
1
answer
52
views
Do GANs have constant running time?
After the model is trained, you just need to input random noise and the generator will output an image, does this mean GANs have constant running time ? I'm asking about both naïve GAN and variants of ...
- 51
3
votes
0
answers
81
views
What are the advantages of GANs over Diffusion Models in image generation?
Diffusion Models have recently gained popularity in the field of image generation, with widely used products such as Stable Diffusion employing this approach and yielding impressive results. While ...
- 51
1
vote
0
answers
101
views
Question about the Conditioning Augmentation technique?
In the paper StackGAN: Text to Photo-realistic Image Synthesis with Stacked Generative Adversarial Networks, the goal is to convert text descriptions into images. The text encoder encodes the ...
- 51
1
vote
2
answers
150
views
The training process of a conditional GAN
For example, consider a dataset like MNIST. I give the conditional vector to produce only the number $7$ for both the generator and discriminator. In the following scenarios, what will the ...
- 15
0
votes
1
answer
111
views
In the conditional GAN (cGAN) architecture, why does the discriminator need conditional variable?
I'm reading about conditional GAN (cGAN) architecture, what I know is that the generator creates images combining both noise vector and conditional variable, the noise vector brings in random elements ...
- 15
0
votes
1
answer
22
views
Generator loss not decreasing while training GAN
I’ve been attempting to create a basic GAN to generate images using this database of flowers (https://www.robots.ox.ac.uk/~vgg/data/flowers/102/).
I’ve spent a few days on this, and largely based my ...
0
votes
0
answers
17
views
Is synthetic data just a placebo for immature models?
I apologize for the provocative question, but let me elaborate.
I am trying to wrap my head around the logic of synthetic data.
When you train a model what you are trying to do is to teach the ground ...
- 295
0
votes
1
answer
37
views
Do WGAN gradients require multi-variable calculus?
The generator tries to maximise this function D(G(z)). That much I understand.
But how can the critic maximise D(x) - D(G(z)). ...
- 13
0
votes
1
answer
37
views
How to deal with sparse 1D data with WGAN-gp
I want to modify WGAN-gp so that it will work on very sparse 1D data (FFT of Gauss signal). Can you suggest any methods or papers that will be helpful?
I have a working WGAN-gp, but it is not ...
1
vote
1
answer
49
views
How the generator loss works in a GAN
I've been reading about GANs so I can implement a simple image generator. It seems that the loss for the generator is given by the following equation: ...
- 13
0
votes
0
answers
16
views
FastGAN implementation has redundant SpectralNorm followed by BatchNorm?
I am implementing a version of FastGAN, and it seems like (see the official repo) they use a Spectral norm directly followed by a Batch norm. Doesn't the spectral norm get fully cancelled by the ...
- 21
0
votes
0
answers
32
views
How to generate quality synthetic images of human subjects to be used for training stable diffusion
I'm trying to generate some synthetic images of humans, that are "good enough" to be used to fine-tune a stable diffusion model, but i'm not sure if this is possible.
I have experimented ...
0
votes
0
answers
19
views
Method of encoding vector graphic data for a GAN
I am working with vector data that I want to train a GAN on. The vector data is a combination of the following forms of vector primitives:
Lines - format: ...
- 143
0
votes
1
answer
95
views
Math behind Diffusion models explanation?
I am recently reading this wonderful article https://lilianweng.github.io/posts/2021-07-11-diffusion-models/#what-are-diffusion-models about the math behind the diffusion models ,
As i dont have a ...
- 3
2
votes
1
answer
75
views
how the GAN architecture maintain similar images close in the latent space?
I am learning about generative models, and I don't quite understand how the GAN architecture can maintain similar generated images close in the latent space. For example, an autoencoder and a ...
- 453
3
votes
2
answers
625
views
Implementing a GAN with control over the output class
I am trying to accomplish the reverse of the typical MNIST in machine learning using a GAN - instead of predicting a number from an image of a digit, I want to reconstruct an image of a digit from a ...
- 143
0
votes
2
answers
56
views
Are popular songs considered outliers to AI
I was thinking about making an adversarial network to generate popular music, so one AI which generates and then two others which detect whether the song is AI generated and the expected view count of ...
0
votes
0
answers
34
views
StyleGAN runtime phenomenom
I was playing around with MobileStyleGAN pretrained model and multithreading and came along with an interesting phenomenom. After a while application is running MobileStyleGAN starts to produce video ...
- 115
1
vote
1
answer
55
views
Comparison of the two alternative forms for the KL divergence [closed]
On page 468 of 'Pattern Recognition and Machine Learning', what does 'the same variables given by the product of two independent univariate Gaussian distributions' mean?
The PDF says,
The green ...
0
votes
0
answers
15
views
Quantization Parameters when converting Quantized Transposed Convolution to Conv2D
A simple way to compute TransposedConv2d is to convert it to a regular Conv2d by padding the input value with zeros, as is described in A guide to convolution arithmetic for deep
learning. Does this ...
- 3
0
votes
1
answer
52
views
Why does the critic of WGAN-GP run more steps than the generator?
As noted in the paper introducing WGAN-GP (see the pseudo-code), for each minibatch of data, the generator's weights are updated only once, and the critic (or discriminator) is updated multiple times. ...
0
votes
0
answers
57
views
Unstable Progressive GAN Training with Exploding Gradients when Fading in New Layers
I'm trying to implement and train a Progressive GAN for TensorFlow (to generate faces like the CelebA dataset), following along with the original paper (https://arxiv.org/pdf/1710.10196.pdf), as well ...
- 1
1
vote
1
answer
73
views
What are alternatives to Inception Score? Can it be used for non-photographic image types?
Most online sources recommend using versions of the Inception score to evaluate the synthetic images generated by a GAN. These scores are pre-trained on the InceptionV3 model. Does this mean that ...
- 11
1
vote
1
answer
51
views
Which face filter algorithms can work on CPU or integrated GPU?
I see many realtime face swap filters and appearance enhancement filters on smartphone apps. Even apps that can make you look like a granny or show you having a frown, no matter what your actual ...
- 113
0
votes
0
answers
34
views
What could be the reason for getting low contrast in GAN generated images?
I am trying to use DCGAN for the waterbirds dataset with the following hyperparameters, but I am getting low-contrast images. It would be great if anyone has any suggestions as to what could be some ...
- 1
0
votes
1
answer
32
views
Query regarding the minimax value function of GANs
In the book Generative AI with Python and TensorFlow 2 from Babcock and Bali (page 172), it is stated that the value function of a GAN is the following:
where D(x) is the output of the discriminator ...
0
votes
1
answer
70
views
References for synthetic images generation from small datasets (~10-50 images)
I'm looking a references (papers / works) for synthetic image generation from small datasets. By small dataset, I mean 10-50 images.
I assume, that the best approaches should be based GAN (cGAN ?) or ...
- 101
-2
votes
1
answer
52
views
I'd like some suggestions on how to effectively master machine learning. I'm new to this and I'm kinda lost [closed]
Thing is, I got a lot of resources to learn from and I have no idea which ones will be the most effective, and where to start. I've come across 2 kinds of books, the ones which talk about the ...
- 1
0
votes
0
answers
148
views
Face variants from a unique artificial human face? (Person for presentation)
If an AI human face generator produces a pretty picture, is there a way to use it as a base for GANs to generate more variants from only that one face in different life conditions?
Furthermore: using ...
0
votes
1
answer
109
views
Keypoint generation in 3D point clouds with Deep Learning
I have a huge dataset of 3D point clouds (each point consists of X,Y,Z coordinates) and another dataset with keypoints (also X,Y,Z) which lie on quite recognizable structures in the point cloud. As a ...
- 1
0
votes
0
answers
34
views
Mean and std in vaerational autoencoder
Are mean and standard deviation in variational autoencoders equal?
If not, then why are both calculated in the same way?
0
votes
1
answer
266
views
How to prevent vanishing/exploding gradients in a GAN with large mini-batch size?
I am training several different GAN architectures, and I noticed that larger batch sizes may lead to vanishing or exploding gradients. In the interest of accelerating training, however, larger batch ...
0
votes
0
answers
14
views
Which deep learning models are suitable for network_parameters-to-images mapping?
I face the problem of learning a mapping (or a translation) $f: (x,\theta) \to x^\prime$, where $x, x^\prime$ are images, $\theta$ is the parameters of a neural network.
I know the models for ...
- 111
1
vote
2
answers
760
views
Converting RGB images to Thermal Images
I am working on a project where I am planning to convert RGB images to thermal images. I can convert to either near infrared spectrum images or far infrared spectrum image.
I am planing on using ...
0
votes
0
answers
49
views
Progressive GAN breaks at higher resolution because of batch size?
I'm trying to reproduce the paper of progressive GANs, by following a tutorial available on github.
All is well up until a resolution of 64:
Then at the resolution of 128 all is going downhill pretty ...
- 111
1
vote
0
answers
130
views
How can I solve the blurring problem in GAN generated images?
In my project I work in dresses dataset. I can solve the problem of black pixels but blurring still existed. I tried many computer vision filters like median filter, Biliteral filter, Sharpen methods ...
2
votes
0
answers
326
views
GAN : Why does a perfect discriminator mean no gradient for the generator?
In the training of a Generative Adversarial Networks (GAN) system, a perfect discriminator (D) is one which outputs 1 ("true image") for all images of the training dataset and 0 ("false ...
- 221
0
votes
0
answers
17
views
Idea for generating time series with irregular time-intervals with GANs
I want to model time-series with irregular time-intervals using GANs. Think of the following (short) data sample
...
- 2,103
-1
votes
1
answer
77
views
Combination of VAE with GAN
I am going to implement a lecture which it aims to generate new images. It uses a variational autoencoder to produce latent vector and then feed it to a gan network as input. My question is, in ...
1
vote
0
answers
136
views
Is it possible to combine DDPM with GAN?
From what I understand in GAN, the main idea is that you have a generator and a discriminator network that are "competing" with each other. The generator trying to make images that the ...
0
votes
1
answer
280
views
Using GANs to generate data augmentations for YOLOv5
I was building a YOLOv5 object detection model, and was looking into researching synthetic methods like GANs to increase the size of my training set in an unsupervised manner.
I know that few-shot ...
- 3
0
votes
1
answer
31
views
Can any GAN's utilize labels in their datasets while they are training?
It seems to me that the Generative Adversarial Networks have a practical issue when trying to reproduce some of their output images
For example, as you can see https://www.youtube.com/watch?v=...
- 101
0
votes
0
answers
13
views
How does the 'moment's matching trick' in GAN training improve the diversity of the generated samples?
I was investigating the TimeGAN code, when i stumbled across the 'moments loss' component. In one of the issues, the author states that this is a 'moment's matching trick' used 'to improve the ...
- 2,103
0
votes
1
answer
62
views
Best way to generate a human face over a face generated by FaceFormer framework?
FaceFormer framework generates a talking face from audio, focusing on the lip and face movement when a person talks. Now from that what would be the best way to generate a human face on top of that?
I ...
- 208
1
vote
2
answers
542
views
GAN with multiple discriminators
I am looking for literature recommendations regarding GANs with multiple discriminators.
In particular, I am looking for examples where each discriminator has a slightly different learning objective, ...
0
votes
0
answers
38
views
Why would one still use a traditional GAN architecture or WGAN architecture instead of a WGAN-GP architecture?
I've been diving into the literature of GANs, and quite early on, I was pretty convinced that WGAN-GPs were the way to go. The WGAN-GP architecture is, as far as I know, theoretically and empirically ...
- 2,103
0
votes
0
answers
15
views
Adding MNIST images by using them as channel inputs
I'm trying to create a generative neural network that can offer "basic sum" mathematical solutions using the MNIST dataset from a conditional input.
I've curated a dataset of MNIST examples ...
- 1
1
vote
2
answers
280
views
Why don't OpenAI train a deep learning model to identify correct and incorrect information in ChatGPT's responses?
I'll preface this by saying that I have little experience in artificial intelligence, so this might be a naive question.
However, in light of the recent controversy surrounding ChatGPT's inability to ...
- 119
0
votes
1
answer
167
views
Why does data augmentation using synthetic data generated by one model improve the performance of another model?
I understand from articles like this one that synthetic data generated by one model based on real data can improve the performance of a second model. Can anyone help me understand the intuition behind ...
