← Back to work
04 · Research

Self-Supervised Semantic Segmentation

Learning Visual Representations Without Labels

Overview

Designed a self-supervised segmentation pipeline for a single high-resolution image, combining CNN patch representations, spatial coordinates and K-means clustering without pretrained models.

Results
0
Labels used
1
High-res source image
Stack
PyTorchCNNContrastive LearningK-means
Notes

Add your write-up here.