ContainersNVIDIADeepSeek-R1NVIDIALlama-3.1-Nemotron-70B-InstructNVIDIAPyTorch
CollectionsNVIDIAOmniverse Kit (FB)NVIDIADeepStream SDKNVIDIAOmniverse Kit App Streaming
ModelsNVIDIAStyleGAN3 pretrained modelsNVIDIAPeopleNetNVIDIATrafficCamNet
ResourcesNVIDIAKit SDK - Windows (PB25h1)NVIDIAKit SDK - Linux (PB25h1)NVIDIARiva Skills Quick Start
Helm ChartsNVIDIAGPU OperatorNVIDIANVIDIA NIM Operator
NVIDIA
NVIDIA
Action Recognition Net
Model
NVIDIA
NVIDIA
Action Recognition Net

5 class action recognition network to recognize what people do in an image.

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):Recognizing actions of people.
Model Type:Action Recognition
Intended Users:An Action Recognition model is intended to benefit developers, researchers, and organizations working on video analysis and surveillance, healthcare providers monitoring patient activities, sports analysts studying player movements, and industries requiring automated systems to understand and respond to human actions, such as autonomous vehicles and interactive robots.
Output:Action Labels
Describe how the model works:Predicts action of person in an image
Technical Limitations:NVIDIA ActionRecognitionNet is trained on HMDB5 which is a subset of HMDB51 containing 1024 videos in total. So it is expected the accuracy of the model on videos other than those from HMDB5 is not at the same level as the number reported in performance section. Model needs to be transferred learned before deployment.
Verified to have met prescribed NVIDIA standards:Yes
Performance Metrics:Top-1 : Precision
Potential Known Risks:This model may fail to detect actions.
Licensing:https://www.nvidia.com/en-us/data-center/products/nvidia-ai-enterprise/eula/