Neural Ensemble Segmentation Suite (NESS), a Python tool designed for semantic segmentation tasks, especially tailored for analyzing fluorescence images of neurons. Currently in development, NESS simplifies segmentation by combining traditional Convolutional Neural Network (CNN) models with state-of-the-art Vision Transformer (ViT) architectures.
Built on popular libraries like keras-unet-collection, segmentation models, and Hugging Face's vision transformers, NESS will be providing a straightforward toolkit for harnessing ensemble methodology to foster fostering synergies that amplify the overall segmentation performance, with an anticipated release in early 2024.