GitHub - DeconvFFT/unsupervised-scene-segmentation: I explore and compare different techniques for unsupervised scene segmentation. I try to answer these research questions: 1.) Can unsupervised convolutional neural networks learn enough structure from data to generate good quality segments? 2.) Is spatial continuity important to generate good quality clusters? 3.) Can we improve results from CNN and GMMs using K-means?

DeconvFFT / unsupervised-scene-segmentation Public

Notifications You must be signed in to change notification settings
Fork 0
Star 1

I explore and compare different techniques for unsupervised scene segmentation. I try to answer these research questions: 1.) Can unsupervised convolutional neural networks learn enough structure from data to generate good quality segments? 2.) Is spatial continuity important to generate good quality clusters? 3.) Can we improve results from CNN…

1 star 0 forks Branches Tags Activity

Star

Notifications

Branches Tags

Name		Name	Last commit message	Last commit date
Latest commit History 2 Commits
AML_K_Means_.ipynb		AML_K_Means_.ipynb
AML_ppt_pdf_version.pdf		AML_ppt_pdf_version.pdf
AML_proj_K_Means_.ipynb		AML_proj_K_Means_.ipynb
CNN_segmentation.ipynb		CNN_segmentation.ipynb
Final_Report_AML.pdf		Final_Report_AML.pdf
Project-template.docx		Project-template.docx
README.md		README.md
Unsupervised_Image_segmentation.ipynb		Unsupervised_Image_segmentation.ipynb
dog_test_img_Segm.jpg		dog_test_img_Segm.jpg