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
    Bi3D Proximity Segmentation
    Model
    NVIDIA
    NVIDIA
    Bi3D Proximity Segmentation

    Bi3D is a binary depth classification network that is used to classify the depth of objects at a given distance.

  • 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):Learning quantized/discretized relative object depths for robotics and autonomous machines.
    Model Type:Bi3D 3D depth estimation.
    Intended Users:Developers building and/or customizing robotics applications.
    Output:Qualtized depth map planes with binary pixel labels.
    Describe how the model works:Network performs binary depth classification from a stereo camera.
    Technical Limitations:Bi3D was designed to be a fast depth predictor at the expense of accuracy. Hence it can only predict relative object depth and not actual depth.
    Performance Metrics:Binary Cross-Entropy.
    Potential Known Risks:N/A
    Licensing:https://developer.download.nvidia.com/licenses/tao_toolkit_21-08_models_eula.pdf