Attention Mechanism for Contrastive Learning in GAN-based Image-to-Image Translation

Hanzhen Zhang, Liguo Zhou, Ruining Wang, Alois Knoll
2/23/2023
cs.CVcs.LGeess.IV

No confident code implementation found for this paper.

Abstract

Using real road testing to optimize autonomous driving algorithms is time-consuming and capital-intensive. To solve this problem, we propose a GAN-based model that is capable of generating high-quality images across different domains. We further leverage Contrastive Learning to train the model in a self-supervised way using image data acquired in the real world using real sensors and simulated images from 3D games. In this paper, we also apply an Attention Mechanism module to emphasize features that contain more information about the source domain according to their measurement of significance. Finally, the generated images are used as datasets to train neural networks to perform a variety of downstream tasks to verify that the approach can fill in the gaps between the virtual and real worlds.

View on arXivView PDF

Code Implementations

No confident code match yet

We couldn't find an author-owned or strongly-evidenced community implementation for this paper. Any repos shown below are weak matches — verify before relying on them.

No code implementations found yet.

Know of an implementation? Let us know in the comments below!

Similar Papers

E$^{2}$GAN: Efficient Training of Efficient GANs for Image-to-Image Translation

81% match2024

HyPER-GAN: Hybrid Patch-Based Image-to-Image Translation for Real-Time Photorealism Enhancement

81% match2026

Efficient generative adversarial networks using linear additive-attention Transformers

81% match2 repos92024

SyntStereo2Real: Edge-Aware GAN for Remote Sensing Image-to-Image Translation while Maintaining Stereo Constraint

80% match2024

Generative Adversarial Networks Bridging Art and Machine Intelligence

80% match2025

An Examination of Offline-Trained Encoders in Vision-Based Deep Reinforcement Learning for Autonomous Driving

80% match2024

Cite this paper

@article{zhang2023attention,
  title  = {Attention Mechanism for Contrastive Learning in GAN-based Image-to-Image Translation},
  author = {Hanzhen Zhang and Liguo Zhou and Ruining Wang and Alois Knoll},
  year   = {2023},
  eprint = {2302.12052},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2302.12052v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2302.12052.svg)](https://codeofpaper.com/paper/2302.12052)

Discussion