Spatial Broadcast Decoder is an architecture that aims to improve disentangling, reconstruction accuracy, and generalization to held-out regions in data space. It provides a particularly dramatic benefit when applied to datasets with small objects.
Source: Watters et al.
Image source: Watters et al.
Source: Spatial Broadcast Decoder: A Simple Architecture for Learning Disentangled Representations in VAEsPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Decoder | 2 | 5.88% |
BIG-bench Machine Learning | 2 | 5.88% |
Semantic Segmentation | 2 | 5.88% |
Object Discovery | 2 | 5.88% |
Unsupervised Object Segmentation | 2 | 5.88% |
blind source separation | 1 | 2.94% |
Speech Separation | 1 | 2.94% |
Clustering | 1 | 2.94% |
Superpixels | 1 | 2.94% |
Component | Type |
|
---|---|---|
🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |