SCAN automatically groups images into semantically meaningful clusters when ground-truth annotations are absent. SCAN is a two-step approach where feature learning and clustering are decoupled. First, a self-supervised task is employed to obtain semantically meaningful features. Second, the obtained features are used as a prior in a learnable clustering approach.
Image source: Gansbeke et al.
Source: SCAN: Learning to Classify Images without LabelsPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Computed Tomography (CT) | 1 | 8.33% |
COVID-19 Diagnosis | 1 | 8.33% |
Decision Making | 1 | 8.33% |
Graph Representation Learning | 1 | 8.33% |
De-identification | 1 | 8.33% |
Classification | 1 | 8.33% |
Clustering | 1 | 8.33% |
General Classification | 1 | 8.33% |
Image Classification | 1 | 8.33% |
Component | Type |
|
---|---|---|
🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |