About:

This module implements an latent space quantization model that generates an enhanced z-stack given a noisy input of Arabidopsis thaliana.

The model is developed to reconstruct an enhanced verison of the noisy input z-stack.

Deep Quantized Latent Representation (DQLR) framework for enhancing z−stack imaging in SAM of Arabidopsis thaliana. We apply quantization in the latent space of the noisy z−stack for enhanced reconstruction. See more details in the paper, Deep Quantized Representation for Enhanced Reconstruction..

Input

Tiff file of a z-stack.

Output

Visually enhanced z-stack.

Trained deep model

Currently we have a pretrained models on image size 256x256.