README.md

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

---

![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/centernet_2d/web-assets/model_demo.png)

# CenterNet-2D: Optimized for Qualcomm Devices

CenterNet-2D is machine learning model that detects objects by finding their center points.

This is based on the implementation of CenterNet-2D found [here](https://github.com/xingyizhou/CenterNet).
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/centernet_2d) 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 |
|---|---|---|---|---|
| QNN_CONTEXT_BINARY | float | qualcomm_qcs8450_proxy | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/centernet_2d/releases/v0.50.2/centernet_2d-qnn_context_binary-float-qualcomm_qcs8450_proxy.zip)
| QNN_CONTEXT_BINARY | float | qualcomm_qcs8550_proxy | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/centernet_2d/releases/v0.50.2/centernet_2d-qnn_context_binary-float-qualcomm_qcs8550_proxy.zip)
| QNN_CONTEXT_BINARY | float | qualcomm_qcs9075 | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/centernet_2d/releases/v0.50.2/centernet_2d-qnn_context_binary-float-qualcomm_qcs9075.zip)
| QNN_CONTEXT_BINARY | float | qualcomm_sa7255p | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/centernet_2d/releases/v0.50.2/centernet_2d-qnn_context_binary-float-qualcomm_sa7255p.zip)
| QNN_CONTEXT_BINARY | float | qualcomm_sa8295p | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/centernet_2d/releases/v0.50.2/centernet_2d-qnn_context_binary-float-qualcomm_sa8295p.zip)
| QNN_CONTEXT_BINARY | float | qualcomm_sa8775p | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/centernet_2d/releases/v0.50.2/centernet_2d-qnn_context_binary-float-qualcomm_sa8775p.zip)
| QNN_CONTEXT_BINARY | float | qualcomm_snapdragon_8_elite_for_galaxy | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/centernet_2d/releases/v0.50.2/centernet_2d-qnn_context_binary-float-qualcomm_snapdragon_8_elite_for_galaxy.zip)
| QNN_CONTEXT_BINARY | float | qualcomm_snapdragon_8_elite_gen5 | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/centernet_2d/releases/v0.50.2/centernet_2d-qnn_context_binary-float-qualcomm_snapdragon_8_elite_gen5.zip)
| QNN_CONTEXT_BINARY | float | qualcomm_snapdragon_8gen3 | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/centernet_2d/releases/v0.50.2/centernet_2d-qnn_context_binary-float-qualcomm_snapdragon_8gen3.zip)
| QNN_CONTEXT_BINARY | float | qualcomm_snapdragon_x2_elite | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/centernet_2d/releases/v0.50.2/centernet_2d-qnn_context_binary-float-qualcomm_snapdragon_x2_elite.zip)
| QNN_CONTEXT_BINARY | float | qualcomm_snapdragon_x_elite | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/centernet_2d/releases/v0.50.2/centernet_2d-qnn_context_binary-float-qualcomm_snapdragon_x_elite.zip)

For more device-specific assets and performance metrics, visit **[CenterNet-2D on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/centernet_2d)**.


### 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/centernet_2d) 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 [CenterNet-2D on GitHub](https://github.com/qualcomm/ai-hub-models/blob/main/src/qai_hub_models/models/centernet_2d) for usage instructions.

## Model Details

**Model Type:** Model_use_case.object_detection

**Model Stats:**
- Model checkpoint: ctdet_coco_dla_2x.pth
- Input resolution: 1 x 3 x 512 x 512
- Number of parameters: 20.2M
- Model size: 37.6 MB

## Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|---|---|---|---|---|---|---
| CenterNet-2D | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 241.143 ms | 17 - 27 MB | NPU
| CenterNet-2D | PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | 269.911 ms | 52 - 52 MB | NPU
| CenterNet-2D | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | 460.051 ms | 54 - 54 MB | NPU
| CenterNet-2D | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | 305.797 ms | 17 - 23 MB | NPU
| CenterNet-2D | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS8550 (Proxy) | 449.131 ms | 1 - 63 MB | NPU
| CenterNet-2D | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS9075 | 462.526 ms | 10 - 15 MB | NPU
| CenterNet-2D | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 310.192 ms | 14 - 21 MB | NPU
| CenterNet-2D | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | 241.806 ms | 3 - 13 MB | NPU
| CenterNet-2D | QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | 254.094 ms | 3 - 3 MB | NPU
| CenterNet-2D | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | 443.359 ms | 3 - 3 MB | NPU
| CenterNet-2D | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | 331.311 ms | 3 - 11 MB | NPU
| CenterNet-2D | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8275 (Proxy) | 584.433 ms | 0 - 9 MB | NPU
| CenterNet-2D | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8550 (Proxy) | 444.163 ms | 4 - 5 MB | NPU
| CenterNet-2D | QNN_CONTEXT_BINARY | float | Qualcomm® SA8775P | 478.365 ms | 0 - 9 MB | NPU
| CenterNet-2D | QNN_CONTEXT_BINARY | float | Qualcomm® QCS9075 | 467.416 ms | 3 - 13 MB | NPU
| CenterNet-2D | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8450 (Proxy) | 597.309 ms | 3 - 13 MB | NPU
| CenterNet-2D | QNN_CONTEXT_BINARY | float | Qualcomm® SA7255P | 584.433 ms | 0 - 9 MB | NPU
| CenterNet-2D | QNN_CONTEXT_BINARY | float | Qualcomm® SA8295P | 501.228 ms | 0 - 5 MB | NPU
| CenterNet-2D | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite For Galaxy Mobile | 307.254 ms | 0 - 9 MB | NPU

## License
* The license for the original implementation of CenterNet-2D can be found
  [here](https://github.com/xingyizhou/CenterNet/blob/master/LICENSE).

## References
* [Objects as Points](https://arxiv.org/abs/1904.07850)
* [Source Model Implementation](https://github.com/xingyizhou/CenterNet)

## 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).