Researchers have developed a deep learning model for in-focus object segmentation in 3D widefield microscopy. By using weak labels generated algorithmically, the model was trained to detect in-focus pixels in widefield microscopy images. The model was fine-tuned using manually segmented ground truth data. The results showed that the model successfully separated in-focus and out-of-focus parts of each focal slice, enabling precise 3D reconstruction of specimens. This approach makes in vivo 3D imaging more accessible and affordable for laboratories with limited resources.
Facebook Research has released the Segment Anything Model (SAM) under the Apache 2.0 license, which uses machine learning to reliably figure out which pixels in an image belong to an object. SAM has been trained on a huge dataset of high-quality images and masks, making it very effective at what it does. Once an image is segmented, those masks can be used to interface with other systems like object detection and other computer vision applications.
Meta has introduced an AI model called the Segment Anything Model (SAM) that can identify individual objects in images and videos, even those not encountered during training. SAM is an image segmentation model that can respond to text prompts or user clicks to isolate specific objects within an image. Meta hopes to "democratize" the process of creating accurate segmentation models by reducing the need for specialized training and expertise. Meta has also assembled a dataset it calls "SA-1B" that includes 11 million images licensed from "a large photo company" and 1.1 billion segmentation masks produced by its segmentation model.
