--- title: CenterPoint canonical_url: "https://www.modelscope.cn/models/Qualcomm-AI-Hub/CenterPoint" md_url: "https://www.modelscope.cn/models/Qualcomm-AI-Hub/CenterPoint.md" repository: Qualcomm-AI-Hub/CenterPoint last_updated: 2026-10-16 license: mit downloads: 0 stars: 0 tags: - android --- # CenterPoint > CenterPoint - Qualcomm-AI-Hub 在 ModelScope 开源的模型。CenterPoint: Optimized for Qualcomm Devices Qualcomm-AI-Hub/CenterPoint 是 ModelScope 魔搭社区上的机器学习模型，采用 mit 许可。 - **Repository**: Qualcomm-AI-Hub/CenterPoint - **License**: mit - **Tags**: android - **Downloads**: 0 - **Stars**: 0 - **Last updated**: 2026-10-16 Source: https://www.modelscope.cn/models/Qualcomm-AI-Hub/CenterPoint --- ![](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/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-cn/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.cn) 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 | |---|---|---|---|---| | ONNX | float | Universal | QAIRT 2.45, ONNX Runtime 1.27.1 | [Download](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/centerpoint/releases/v0.62.1/centerpoint-onnx-float.zip) | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/centerpoint/releases/v0.62.1/centerpoint-qnn_dlc-float.zip) | TFLITE | float | Universal | | [Download](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/centerpoint/releases/v0.62.1/centerpoint-tflite-float.zip) For more device-specific assets and performance metrics, visit **[CenterPoint on Qualcomm® AI Hub](https://aihub.qualcomm.cn/models/centerpoint)**. ### Option 2: Export with Custom Configurations Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models-cn/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-cn/blob/main/src/qai_hub_models/models/centerpoint) for usage instructions. ## Model Details **Model Type:** Model_use_case.driver_assistance **Model Stats:** - Input resolution: 5x20x5, 5x4, 5 - Model checkpoint: PointPillars - Model size: 83.3 MB - Number of parameters: 21.8M ## Performance Summary | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |---|---|---|---|---|---|--- | CenterPoint | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 318.743 ms | 1057 - 1953 MB | NPU | CenterPoint | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 409.496 ms | 1049 - 1069 MB | NPU | CenterPoint | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 299.933 ms | 1044 - 1417 MB | NPU | CenterPoint | ONNX | float | Snapdragon® 8 Elite Mobile | 299.933 ms | 1044 - 1417 MB | NPU | CenterPoint | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 251.197 ms | 1056 - 1446 MB | NPU | CenterPoint | QNN_DLC | float | Snapdragon® X2 Elite | 280.558 ms | 2 - 2 MB | NPU | CenterPoint | QNN_DLC | float | Snapdragon® X Elite | 416.365 ms | 2 - 2 MB | NPU | CenterPoint | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 308.26 ms | 0 - 690 MB | NPU | CenterPoint | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 885.617 ms | 1 - 552 MB | NPU | CenterPoint | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 455.059 ms | 2 - 11 MB | NPU | CenterPoint | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 421.579 ms | 3 - 5 MB | NPU | CenterPoint | QNN_DLC | float | Qualcomm® SA8775P | 480.156 ms | 1 - 407 MB | NPU | CenterPoint | QNN_DLC | float | Qualcomm® SA8650P | 480.156 ms | 1 - 407 MB | NPU | CenterPoint | QNN_DLC | float | Qualcomm® SA8255P | 480.156 ms | 1 - 407 MB | NPU | CenterPoint | QNN_DLC | float | Qualcomm® QCS8450 | 885.617 ms | 1 - 552 MB | NPU | CenterPoint | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 472.537 ms | 2 - 11 MB | NPU | CenterPoint | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 416.365 ms | 2 - 2 MB | NPU | CenterPoint | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 297.13 ms | 0 - 419 MB | NPU | CenterPoint | QNN_DLC | float | Qualcomm® SA7255P | 1025.144 ms | 2 - 412 MB | NPU | CenterPoint | QNN_DLC | float | Qualcomm® SA8295P | 530.742 ms | 0 - 403 MB | NPU | CenterPoint | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 297.13 ms | 0 - 419 MB | NPU | CenterPoint | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 253.537 ms | 2 - 413 MB | NPU | CenterPoint | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 4314.218 ms | 1859 - 1868 MB | CPU | CenterPoint | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 5485.207 ms | 1840 - 1851 MB | CPU | CenterPoint | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 6081.167 ms | 2361 - 2383 MB | CPU | CenterPoint | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 4674.031 ms | 1877 - 1879 MB | CPU | CenterPoint | TFLITE | float | Qualcomm® SA8775P | 5497.631 ms | 1809 - 1815 MB | CPU | CenterPoint | TFLITE | float | Qualcomm® SA8650P | 5497.631 ms | 1809 - 1815 MB | CPU | CenterPoint | TFLITE | float | Qualcomm® SA8255P | 5497.631 ms | 1809 - 1815 MB | CPU | CenterPoint | TFLITE | float | Qualcomm® QCS8450 | 5485.207 ms | 1840 - 1851 MB | CPU | CenterPoint | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 5241.517 ms | 2363 - 2385 MB | CPU | CenterPoint | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 2767.298 ms | 1809 - 1818 MB | CPU | CenterPoint | TFLITE | float | Qualcomm® SA7255P | 6339.224 ms | 1847 - 1856 MB | CPU | CenterPoint | TFLITE | float | Qualcomm® SA8295P | 3650.458 ms | 1805 - 1811 MB | CPU | CenterPoint | TFLITE | float | Snapdragon® 8 Elite Mobile | 2767.298 ms | 1809 - 1818 MB | CPU | CenterPoint | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 884.006 ms | 1941 - 1949 MB | CPU ## License * The license for the original implementation of CenterPoint can be found [here](https://github.com/tianweiy/CenterPoint/blob/master/LICENSE). ## Community * For questions or feedback, chat with our engineers in [Qualcomm® AI Hub Workbench](https://workbench.aihub.qualcomm.cn).