Hugging Peers

PointNet

☆ Save
by Qualcomm-AI-Hub image-segmentation apache-2.0
0 downloads 0 likes on ModelScope updated

Model card

PointNet: Optimized for Qualcomm Devices

PointNet is a pioneering neural network architecture designed to directly consume unordered point cloud data for tasks such as classification and segmentation. It learns spatial features from raw 3D points without requiring voxelization or image projections.

This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the Qualcomm® AI Hub Models library to export with custom configurations. More details on model performance across various devices, can be found here.

Qualcomm AI Hub Models uses Qualcomm AI Hub Workbench to compile, profile, and evaluate this model. Sign up to run these models on a hosted Qualcomm® device.

Getting Started

There are two ways to deploy this model on your device:

Option 1: Download Pre-Exported Models

Below are pre-exported model assets ready for deployment.

Runtime Precision Chipset SDK Versions Download
ONNX float Universal QAIRT 2.45, ONNX Runtime 1.27.1 Download
QNN_DLC float Universal QAIRT 2.45 Download
TFLITE float Universal QAIRT 2.45 Download

For more device-specific assets and performance metrics, visit PointNet on Qualcomm® AI Hub.

Option 2: Export with Custom Configurations

Use the Qualcomm® AI Hub Models Python library to compile and export the model with your own:

  • Custom weights (e.g., fine-tuned checkpoints)
  • Custom input shapes
  • Target device and runtime configurations

This option is ideal if you need to customize the model beyond the default configuration provided here.

See our repository for PointNet on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.semantic_segmentation

Model Stats:

  • Input resolution: 1x3x1024
  • Model checkpoint: save
  • Model size: 13.2 MB

Performance Summary

Model Runtime Precision Chipset Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit
PointNet ONNX float Snapdragon® X2 Elite 0.302 ms 1 - 1 MB NPU
PointNet ONNX float Snapdragon® X Elite 0.656 ms 7 - 7 MB NPU
PointNet ONNX float Snapdragon® 8 Gen 3 Mobile 0.399 ms 0 - 43 MB NPU
PointNet ONNX float Snapdragon® 8 Gen 1 Mobile 0.869 ms 0 - 48 MB NPU
PointNet ONNX float Qualcomm® Dragonwing™ IQ-8275 0.765 ms 0 - 3 MB NPU
PointNet ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.648 ms 0 - 27 MB NPU
PointNet ONNX float Qualcomm® QCS8450 0.869 ms 0 - 48 MB NPU
PointNet ONNX float Qualcomm® Dragonwing™ IQ-9075 0.804 ms 0 - 3 MB NPU
PointNet ONNX float Qualcomm® Dragonwing™ IQ-X7181 0.656 ms 7 - 7 MB NPU
PointNet ONNX float Qualcomm® Dragonwing™ Q-8750 0.419 ms 0 - 27 MB NPU
PointNet ONNX float Snapdragon® 8 Elite Mobile 0.419 ms 0 - 27 MB NPU
PointNet ONNX float Snapdragon® 8 Elite Gen 5 Mobile 0.318 ms 0 - 23 MB NPU
PointNet QNN_DLC float Snapdragon® X2 Elite 0.439 ms 1 - 1 MB NPU
PointNet QNN_DLC float Snapdragon® X Elite 0.768 ms 1 - 1 MB NPU
PointNet QNN_DLC float Snapdragon® 8 Gen 3 Mobile 0.414 ms 0 - 42 MB NPU
PointNet QNN_DLC float Snapdragon® 8 Gen 1 Mobile 0.861 ms 0 - 47 MB NPU
PointNet QNN_DLC float Qualcomm® Dragonwing™ IQ-8275 0.734 ms 0 - 2 MB NPU
PointNet QNN_DLC float Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.662 ms 0 - 1 MB NPU
PointNet QNN_DLC float Qualcomm® SA8775P 0.907 ms 0 - 25 MB NPU
PointNet QNN_DLC float Qualcomm® SA8650P 0.907 ms 0 - 25 MB NPU
PointNet QNN_DLC float Qualcomm® SA8255P 0.907 ms 0 - 25 MB NPU
PointNet QNN_DLC float Qualcomm® QCS8450 0.861 ms 0 - 47 MB NPU
PointNet QNN_DLC float Qualcomm® Dragonwing™ IQ-9075 0.815 ms 0 - 2 MB NPU
PointNet QNN_DLC float Qualcomm® Dragonwing™ IQ-X7181 0.768 ms 1 - 1 MB NPU
PointNet QNN_DLC float Qualcomm® Dragonwing™ Q-8750 0.48 ms 0 - 23 MB NPU
PointNet QNN_DLC float Qualcomm® SA7255P 1.869 ms 0 - 24 MB NPU
PointNet QNN_DLC float Qualcomm® SA8295P 1.097 ms 0 - 20 MB NPU
PointNet QNN_DLC float Snapdragon® 8 Elite Mobile 0.48 ms 0 - 23 MB NPU
PointNet QNN_DLC float Snapdragon® 8 Elite Gen 5 Mobile 0.32 ms 0 - 25 MB NPU
PointNet TFLITE float Snapdragon® 8 Gen 3 Mobile 0.412 ms 0 - 41 MB NPU
PointNet TFLITE float Snapdragon® 8 Gen 1 Mobile 0.858 ms 0 - 46 MB NPU
PointNet TFLITE float Qualcomm® Dragonwing™ IQ-8275 0.757 ms 0 - 9 MB NPU
PointNet TFLITE float Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.662 ms 0 - 1 MB NPU
PointNet TFLITE float Qualcomm® SA8775P 0.929 ms 0 - 25 MB NPU
PointNet TFLITE float Qualcomm® SA8650P 0.929 ms 0 - 25 MB NPU
PointNet TFLITE float Qualcomm® SA8255P 0.929 ms 0 - 25 MB NPU
PointNet TFLITE float Qualcomm® QCS8450 0.858 ms 0 - 46 MB NPU
PointNet TFLITE float Qualcomm® Dragonwing™ IQ-9075 0.819 ms 0 - 8 MB NPU
PointNet TFLITE float Qualcomm® Dragonwing™ Q-8750 0.483 ms 0 - 28 MB NPU
PointNet TFLITE float Qualcomm® SA7255P 1.88 ms 0 - 24 MB NPU
PointNet TFLITE float Qualcomm® SA8295P 1.125 ms 0 - 21 MB NPU
PointNet TFLITE float Snapdragon® 8 Elite Mobile 0.483 ms 0 - 28 MB NPU
PointNet TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 0.321 ms 0 - 24 MB NPU

License

  • The license for the original implementation of PointNet can be found here.

Community