The dataset contains 8000 annotated images for both liquid and solid materials the vessels containing them in chemistry labs, medical labs, and any area where liquids and solids are handled within containers.
There are two levels of annotation for each image. One annotation set for vessels and the second for the materials inside these vessels.
The segmentation include semantic level segmentation (vessel, liquid, solid, filled, foam...) and instance… See the full description on the dataset page:
https://huggingface.co/datasets/FlyingFrog/LabPics.
