Semantic Image Segmentation

[atanas1054]

trafficstars

Hello,

Can this method be used for semantic image segmentation? I have a dataset of unlabeled real images and a dataset of labeled synthetic images? Will the big resolution of the image be a problem?

[mikigom]

@atanas1054 Please look up about domain adaptation or transfer learning. The method in this repo is fit for use of unsupervised image generation so the approach you said and the simulated generation are not similar in purpose.

However, I think it is of worth to test cross-domain adaptation from generated data. The quality of result will rely upon task and domain.

For more information, I recommend for you to read these papers.

• Domain-Adversarial Training of Neural Networks
• Section 5 of Improved Techniques for Training GANs
• A Survey on Transfer Learning