Models
Models
Currently being reviewed
SAM
Vision transformer that turns your doodles into pixel-perfect masks.
DeepLab
Atrous convolutions for multi-scale context.
CornerNet
Detects objects by their crispy corner keypoints.
UNet
The contracting-expanding architecture that turned biomedical imaging upside down.
SegNet
The OG encoder-decoder for pixel-wise labeling way before it was cool.
ENet
Packs a punch on semantic segmentation without heavyweight computations.
Last updated on