Yes, we GAN: Applying Adversarial Techniques for Autonomous Driving

By Michal Uřičář, Pavel Krízek, David Hurych, Ibrahim Sobh, Senthil Kumar Yogamani, Patrick Denny. 2019

60 citations recorded by the source corpus.

Generative Adversarial Networks (GAN) have gained a lot of popularity from their introduction in 2014 till present. Research on GAN is rapidly growing and there are many variants of the original GAN focusing on various aspects of deep learning. GAN are perceived as the most impactful direction of machine learning in the last decade. This paper focuses on the application of GAN in autonomous driving including topics such as advanced data augmentation, loss function learning, semi-supervised learning, etc. We formalize and review key applications of adversarial techniques and discuss challenges and open problems to be addressed.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Generative Models, Robotics & Control

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Ciarán Hughes
2. Ibrahim Sobh
3. Divya Saxena
4. Jiannong Cao
5. B. R. Kiran
6. Patrick Pérez
7. Bourahla Omar
8. Karl Amende

In Computer Vision and Pattern Recognition

1. Xi Li
2. Senthil Kumar Yogamani
3. Abdul Jabbar
4. Bourahla Omar
5. Ganesh Sistu
6. Stefan Milz
7. Jiannong Cao
8. Sumanth Chennupati

In Generative Models

1. Ibrahim Sobh
2. Divya Saxena
3. Jiannong Cao
4. Patrick Pérez
5. Bourahla Omar
6. Abdul Jabbar
7. Pavel Krízek
8. Xi Li

In Robotics & Control

1. Ganesh Sistu
2. Michal Uřičář
3. Ibrahim Sobh
4. Patrick Pérez
5. Senthil Kumar Yogamani
6. Patrick Mannion
7. Victor Talpaert

Citations in the corpus

• WoodScape: A Multi-Task, Multi-Camera Fisheye Dataset for Autonomous Driving (2019, background): “The task of soiling detection was to our best knowledge first defined in [56].”
• Deep Reinforcement Learning for Autonomous Driving: A Survey (2020, background): “o avoid this expensive inner loop. In practice, GAIL trains a policy close enough to the expert policy to fool a discriminator. This process is similar to Generative Adversarial Networks (GANs) [74], [75]. The resulting policy must travel the same MDP states as the expert, or the discriminator would”
• A Survey on Generative Adversarial Networks: Variants, Applications, and Training (2020, methodology): “GANs have also been used in variety of other domains such as computer security [175], chess game playing [176], network pruning [177, 178], spatial representation learning [179], mobile user profiling [180], data augmentation [181, 182], heterogeneous information networks [183], privacypreserving [1”
• SoilingNet: Soiling Detection on Automotive Surround-View Cameras (2019, methodology): “To our best knowledge, the soiling detection task was so far only briefly described in our previous paper [14].”
• Generative Adversarial Networks (GANs) (2020, background): “Apart from it, GANs has also shown its potential for several other applications, such as Steganalysis [150], information retrieval [151], spatio-temporal data prediction [152], autonomous driving [153]–[155], speech enhancement [156], etc.”