Unsupervised Cross Domain Semantic Segmentation

The objective is to perform semantic segmentation on the IITK security surveillance dataset in an unsupervised setup using domain adaptation.
Semantic segmentation assigns each pixel a semantic label, e.g., person, car, road or tree, in an image, which can be useful for security purposes.

1. Unsupervised Cross Domain Semantic Segmentation Group:55 Shanu Kumar (150659), Sourabh Kumar (170716)

2. Problem Statement The objective is to perform semantic segmentation on the IITK security surveillance dataset in an unsupervised setup using domain adaptation. Semantic segmentation assigns each pixel a semantic label, e.g., person, car, road or tree, in an image, which can be useful for security purposes.

3. Learning to Adapt Structured Output Space for Semantic Segmentation