v0.54.0

f54beb5 verified 11 days ago

5.56 kB

	---
	library_name: pytorch
	license: other
	tags:
	- android
	pipeline_tag: other

	---

	![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/centerpoint/web-assets/model_demo.png)

	# CenterPoint: Optimized for Qualcomm Devices

	CenterPoint is a LiDAR-based 3D object detection model that detects objects by predicting their centers and regressing other attributes. It is designed for high accuracy and real-time performance in autonomous driving applications.

	This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/src/qai_hub_models/models/centerpoint) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).

	Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.com) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) 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 \|
	\|---\|---\|---\|---\|---\|
	\| TFLITE \| float \| Universal \| \| [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/centerpoint/releases/v0.54.0/centerpoint-tflite-float.zip)

	For more device-specific assets and performance metrics, visit [CenterPoint on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/centerpoint).


	### Option 2: Export with Custom Configurations

	Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/src/qai_hub_models/models/centerpoint) 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 [CenterPoint on GitHub](https://github.com/qualcomm/ai-hub-models/blob/main/src/qai_hub_models/models/centerpoint) for usage instructions.

	## Model Details

	Model Type: Model_use_case.driver_assistance

	Model Stats:
	- Model checkpoint: PointPillars
	- Input resolution: 5x20x5, 5x4, 5
	- Number of parameters: 21.8M
	- Model size: 83.3 MB

	## Performance Summary
	\| Model \| Runtime \| Precision \| Chipset \| Inference Time (ms) \| Peak Memory Range (MB) \| Primary Compute Unit
	\|---\|---\|---\|---\|---\|---\|---
	\| CenterPoint \| QNN_DLC \| float \| Snapdragon® 8 Elite Gen 5 Mobile \| 177.495 ms \| 2 - 717 MB \| NPU
	\| CenterPoint \| QNN_DLC \| float \| Snapdragon® 8 Elite Mobile \| 210.389 ms \| 0 - 461 MB \| NPU
	\| CenterPoint \| QNN_DLC \| float \| Snapdragon® X2 Elite \| 184.021 ms \| 2 - 2 MB \| NPU
	\| CenterPoint \| QNN_DLC \| float \| Snapdragon® X Elite \| 322.285 ms \| 2 - 2 MB \| NPU
	\| CenterPoint \| QNN_DLC \| float \| Snapdragon® X Elite \| 322.285 ms \| 2 - 2 MB \| NPU
	\| CenterPoint \| QNN_DLC \| float \| Snapdragon® 8 Gen 3 Mobile \| 252.132 ms \| 0 - 753 MB \| NPU
	\| CenterPoint \| QNN_DLC \| float \| Qualcomm® QCS8550 (Proxy) \| 330.273 ms \| 2 - 4 MB \| NPU
	\| CenterPoint \| QNN_DLC \| float \| Qualcomm® SA8775P \| 397.327 ms \| 1 - 703 MB \| NPU
	\| CenterPoint \| QNN_DLC \| float \| Qualcomm® SA8775P \| 397.327 ms \| 1 - 703 MB \| NPU
	\| CenterPoint \| QNN_DLC \| float \| Qualcomm® SA8775P \| 397.327 ms \| 1 - 703 MB \| NPU
	\| CenterPoint \| QNN_DLC \| float \| Qualcomm® SA7255P \| 920.506 ms \| 0 - 450 MB \| NPU
	\| CenterPoint \| QNN_DLC \| float \| Qualcomm® SA8295P \| 443.799 ms \| 0 - 449 MB \| NPU
	\| CenterPoint \| QNN_DLC \| float \| Qualcomm® QCS9075 \| 423.343 ms \| 2 - 11 MB \| NPU
	\| CenterPoint \| QNN_DLC \| float \| Snapdragon® 8 Elite For Galaxy Mobile \| 210.389 ms \| 0 - 461 MB \| NPU
	\| CenterPoint \| QNN_DLC \| float \| Qualcomm® QCS8450 (Proxy) \| 522.82 ms \| 2 - 738 MB \| NPU
	\| CenterPoint \| TFLITE \| float \| Snapdragon® 8 Elite Gen 5 Mobile \| 2565.55 ms \| 1866 - 1877 MB \| CPU
	\| CenterPoint \| TFLITE \| float \| Snapdragon® 8 Elite Mobile \| 2623.124 ms \| 1853 - 1862 MB \| CPU
	\| CenterPoint \| TFLITE \| float \| Snapdragon® 8 Gen 3 Mobile \| 4173.433 ms \| 1865 - 1874 MB \| CPU
	\| CenterPoint \| TFLITE \| float \| Qualcomm® QCS8550 (Proxy) \| 4857.49 ms \| 1890 - 1891 MB \| CPU
	\| CenterPoint \| TFLITE \| float \| Qualcomm® SA8775P \| 5384.655 ms \| 1809 - 1815 MB \| CPU
	\| CenterPoint \| TFLITE \| float \| Qualcomm® SA8775P \| 5384.655 ms \| 1809 - 1815 MB \| CPU
	\| CenterPoint \| TFLITE \| float \| Qualcomm® SA8775P \| 5384.655 ms \| 1809 - 1815 MB \| CPU
	\| CenterPoint \| TFLITE \| float \| Qualcomm® SA7255P \| 6211.94 ms \| 1838 - 1846 MB \| CPU
	\| CenterPoint \| TFLITE \| float \| Qualcomm® SA8295P \| 3456.037 ms \| 1807 - 1813 MB \| CPU
	\| CenterPoint \| TFLITE \| float \| Qualcomm® QCS9075 \| 5172.923 ms \| 2364 - 2386 MB \| CPU
	\| CenterPoint \| TFLITE \| float \| Snapdragon® 8 Elite For Galaxy Mobile \| 2623.124 ms \| 1853 - 1862 MB \| CPU
	\| CenterPoint \| TFLITE \| float \| Qualcomm® QCS8450 (Proxy) \| 6894.857 ms \| 1842 - 1852 MB \| CPU

	## License
	* The license for the original implementation of CenterPoint can be found
	[here](https://github.com/tianweiy/CenterPoint/blob/master/LICENSE).



	## Community
	* Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI.
	* For questions or feedback please [reach out to us](mailto:ai-hub-support@qti.qualcomm.com).