GPUTab

Train and use AI image translation models (denoising, segmentation) directly in your browser.

An article about this web implementation is in preparation. In the meantime, if you find this code helpful, please consider citing the precursor work with open-source image-to-image translation code:

Optical tomography in a single camera frame using fringe-encoded deep-learning full-field OCT
Viacheslav Mazlin
Biomedical Optics Express, 2024
DOI
TensorFlow.js WebGPU U-Net Image Processing