https://github.com/theveryhim/segmentation-clustering
Using clustering methods such as K-means as segmentors
https://github.com/theveryhim/segmentation-clustering
clustering deep-learning image-processing image-segmentation unsupervised-learning
Last synced: 11 months ago
JSON representation
Using clustering methods such as K-means as segmentors
- Host: GitHub
- URL: https://github.com/theveryhim/segmentation-clustering
- Owner: theveryhim
- License: mit
- Created: 2025-07-28T18:33:36.000Z (11 months ago)
- Default Branch: main
- Last Pushed: 2025-07-28T18:50:12.000Z (11 months ago)
- Last Synced: 2025-07-28T20:45:36.424Z (11 months ago)
- Topics: clustering, deep-learning, image-processing, image-segmentation, unsupervised-learning
- Language: Jupyter Notebook
- Homepage:
- Size: 2.95 MB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Image Segmentation with Unsupervised and Pretrained Models
Here in this repo we try to segment some images using 3 types of methods:
## Unsupervised Segmentation
Segment all images using the following algorithms:
- **Fuzzy C-Means**
- **K-Means**
## Feature-Based Segmentation
Using a **pretrained ResNet model** (or any other suitable method):
- Extract **feature maps** from the input images.
- Resize these maps using **interpolation** or another technique to match the original image size.
- Concatenate the resized feature maps with the original image as input.
- Apply **Fuzzy C-Means** and **K-Means** segmentation again using this augmented input.
## Segmentation with SAM Model
Use the **pretrained SAM (Segment Anything Model)** to perform image segmentation on the same dataset.
## Results
