Using deep learning to create fake images of previously unseen objects.
Master thesis
Permanent lenke
https://hdl.handle.net/11250/2786171Utgivelsesdato
2021Metadata
Vis full innførselSamlinger
- Studentoppgaver (TN-IDE) [823]
Sammendrag
Deep learning artificial neural networks are implemented in machines at an increasing rate inorder to make them think and act the human way. A popular use of these neural networksis in object detection software, making the machines able to know their environment. Aproblem is that the training of the neural network requires manual effort, in order to createthe training dataset.
Will advances in the training of Generative models, with the introduction of adversarialtraining be able to reduce this manual effort? This thesis explores if (assesses how) the Generative Adversarial Networks (GAN) model can be used to supplement a training datasetwithout the manual registration effort.
The main focus is the Deep Convolutional Generative Adversarial Network (DCGAN) andthe Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP) versions of GANs. The thesis includes the theoretical background and some practical implementation issues for experimenting with these GAN variants.
Three main experiments are presented. First, testing how these generative networks perform on a dataset of images with boxes containing objects with different orientations andpositions. Second, experimenting with only a limited amount of images used for training,before testing two proposed improvement methods, pre-training the models and expandingthe dataset with the generated (artificial) images. For the last experiment the object detection software will be tested on the generated images.
The results show that it is possible to use Generative Adversarial Networks to generate newadditional images for a limited dataset. Perhaps not impressing, but nevertheless, up to4 % of the generated artificial images have the sufficient quality to be considered as newadditions to the dataset.