ContainersNVIDIADeepSeek-R1NVIDIALlama-3.1-Nemotron-70B-InstructNVIDIAPyTorch
CollectionsNVIDIAOmniverse Kit (FB)NVIDIADeepStream SDKNVIDIAOmniverse Kit App Streaming
ModelsNVIDIAStyleGAN3 pretrained modelsNVIDIAPeopleNetNVIDIATrafficCamNet
ResourcesNVIDIARiva Skills Quick Start
Helm ChartsNVIDIAGPU OperatorNVIDIANVIDIA NIM Operator
NVIDIA
NVIDIA
SegIC
Model
NVIDIA
NVIDIA
SegIC

In-context segmentation model trained on commercial data.

This model is backed by NVIDIA's Plus Plus (++) Promise
to learn more about the quality of the datasets used to train this model.
FieldResponse
Intended Application(s) & Domain(s):In-Context Segmentation
Model Type:Semantic Segmentation
Intended Users:This model is intended for developers to build in-context semantic segmentation applications.
Output:Segmentation masks
Describe how the model works:Generates segmentation masks of the test images/frames based on visual prompts features.
Technical Limitations:This model may fail to segment the target items or segment wrong areas.
Verified to have met prescribed NVIDIA standards:Yes
Performance Metrics:Mean Intersection over Union (mIoU) for segmentation evaluation; mean Average Precision (mAP) for processed bounding box evaluation
Licensing:NVIDIA Open Model License