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
NVOF
Model
NVIDIA
NVIDIA
NVOF

NVOF is a deep learning based optical flow estimation and stereo matching solution.

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):Building motion analysis capabilities for robotics and autonomous machind
Model Type:Optical Flow estimation
Intended Users:Developers who need to conduct motion analysis.
Output:Continuous Optical Flow Map
Describe how the model works:The model takes two RGB images as input and produces a dense optical flow image that represents the motion of each pixel between the two frames
Technical Limitations:The model's motion analysis results may not be 100% accurate, particularly in occluded areas, transparent areas. Significant displacement between the two input images, as well as noticeable changes in lighting or noise, can lead to incorrect motion vector outputs.
Verified to have met prescribed NVIDIA standards:Yes
Performance Metrics:AEPE(average end point error)
Potential Known Risks:Imprecise motion analysis results can lead to incorrect recognition and understanding of motion
Licensing:NVIDIA Open Model License Agreement