Open Images Pre-trained Instance Segmentation
Instance segmentation is a popular computer vision technique that can identify each instance of multiple objects in a frame at the pixel level. Instance segmentation will not only produce bounding boxes around the object, but also segmentation masks. This model object contains pretrained weights that may be used as a starting point with the following instance segmentation networks in TAO Toolkit to facilitate transfer learning.
The following instance segmentation architectures are supported in TAO Toolkit.
MaskRCNN
The pre-trained weights are trained on a subset of the Google OpenImages dataset.
The following backbones are supported with these MaskRCNN networks:
resnet10/resnet18/resnet34/resnet50/resnet101
To see the full list of all the backbones, scroll over to the version history tab.
These are unpruned models with just the feature extractor weights, and may not be used without re-training in an instance segmentation application.
For more instructions on downloading and using the models defined here, refer to the NGC catalog page.