The simulation visualises image data flowing down through U-Net's contracting encoder and back up through its expanding decoder, showing the skip connections carrying detailed feature maps across the U in 3D.
Pick a sample input image and press play to animate the forward pass layer by layer, or drag the speed slider and scrub manually to inspect how resolution and feature depth change at each stage.
Image select, speed slider, play/pause, rebuild
The original 2015 U-Net paper won the ISBI cell tracking challenge by a large margin using only about 30 annotated training images, an astonishingly small dataset by deep learning standards.
The simulation visualises image data flowing down through U-Net's contracting encoder and back up through its expanding decoder, showing the skip connections carrying detailed feature maps across the U in 3D.
The simulation visualises image data flowing down through U-Net's contracting encoder and back up through its expanding decoder, showing the skip connections carrying detailed feature maps across the U in 3D.
Pick a sample input image and press play to animate the forward pass layer by layer, or drag the speed slider and scrub manually to inspect how resolution and feature depth change at each stage.
The original 2015 U-Net paper won the ISBI cell tracking challenge by a large margin using only about 30 annotated training images, an astonishingly small dataset by deep learning standards.