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
BEVFusion for 3D Object Detection
Model
NVIDIA
NVIDIA
BEVFusion for 3D Object Detection

BEVFusion model to detect 3D objects from point cloud and RGB data.

1 Version
bevfusion_1.0Selected
08/23/2024 1:02 AM UTC464.53 MBBatch Size: 1GPU: V100
SHA256 digests
KeyValue
tao3d_bevfusion_epoch4.pth9bdadb4777875dc40de7ba4110f643258688ce30438c49edbef4a9e467ff6e98
experiment.yaml683475049003120b1f698ae15b7c734488d44aba024f5cc41514d36fb59ba723
TAO Toolkit API
KeyValue
is_backboneFalse
endpoints['bevfusion']
format_version0.0.1
model_formatpth
tao_version['>=5.5.0']
trainableTrue
TAO Toolkit UI Metadata
KeyValue
licenseNVAIE EULA
task3D Object Detection
format_version0.0.1
backbone_typetransformer
domainpurpose-built
accuracy79.35%
num_parameters157M
backbone_classswin