---
title: MediaPipe-Hand-Detection
canonical_url: "https://www.modelscope.cn/models/Qualcomm-AI-Hub/MediaPipe-Hand-Detection"
md_url: "https://www.modelscope.cn/models/Qualcomm-AI-Hub/MediaPipe-Hand-Detection.md"
repository: Qualcomm-AI-Hub/MediaPipe-Hand-Detection
last_updated: 2026-10-16
license: apache-2.0
downloads: 0
stars: 0
tags:
  - real_time
  - android
---

# MediaPipe-Hand-Detection

> MediaPipe-Hand-Detection - Qualcomm-AI-Hub 在 ModelScope 开源的模型。MediaPipe-Hand-Detection: Optimized for Qualcomm Devices

Qualcomm-AI-Hub/MediaPipe-Hand-Detection 是 ModelScope 魔搭社区上的机器学习模型，采用 apache-2.0 许可。

- **Repository**: Qualcomm-AI-Hub/MediaPipe-Hand-Detection
- **License**: apache-2.0
- **Tags**: real_time, android
- **Downloads**: 0
- **Stars**: 0
- **Last updated**: 2026-10-16

Source: https://www.modelscope.cn/models/Qualcomm-AI-Hub/MediaPipe-Hand-Detection

---

![](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/mediapipe_hand/web-assets/model_demo.png)

# MediaPipe-Hand-Detection: Optimized for Qualcomm Devices

The MediaPipe Hand Landmark Detector is a machine learning pipeline that predicts bounding boxes and pose skeletons of hands in an image.

This is based on the implementation of MediaPipe-Hand-Detection found [here](https://github.com/zmurez/MediaPipePyTorch/).
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/mediapipe_hand) 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/mediapipe_hand/releases/v0.62.1/mediapipe_hand-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/mediapipe_hand/releases/v0.62.1/mediapipe_hand-qnn_dlc-float.zip)
| TFLITE | float | Universal | QAIRT 2.45 | [Download](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/mediapipe_hand/releases/v0.62.1/mediapipe_hand-tflite-float.zip)

For more device-specific assets and performance metrics, visit **[MediaPipe-Hand-Detection on Qualcomm® AI Hub](https://aihub.qualcomm.cn/models/mediapipe_hand)**.


### 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/mediapipe_hand) 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 [MediaPipe-Hand-Detection on GitHub](https://github.com/qualcomm/ai-hub-models-cn/blob/main/src/qai_hub_models/models/mediapipe_hand) for usage instructions.

## Model Details

**Model Type:** Model_use_case.object_detection

**Model Stats:**
- Input resolution: 256x256
- Model size (hand_detector) (float): 6.75 MB
- Model size (hand_landmark_detector) (float): 7.70 MB
- Number of parameters (hand_detector): 1.76M
- Number of parameters (hand_landmark_detector): 2.01M

## Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|---|---|---|---|---|---|---
| hand_detector | ONNX | float | Snapdragon® X2 Elite | 0.487 ms | 2 - 2 MB | NPU
| hand_detector | ONNX | float | Snapdragon® X Elite | 0.851 ms | 0 - 0 MB | NPU
| hand_detector | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 0.592 ms | 0 - 65 MB | NPU
| hand_detector | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 1.5 ms | 1 - 63 MB | NPU
| hand_detector | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 1.39 ms | 1 - 4 MB | NPU
| hand_detector | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 0.837 ms | 0 - 6 MB | NPU
| hand_detector | ONNX | float | Qualcomm® QCS8450 | 1.5 ms | 1 - 63 MB | NPU
| hand_detector | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 1.266 ms | 1 - 3 MB | NPU
| hand_detector | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 0.851 ms | 0 - 0 MB | NPU
| hand_detector | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 0.498 ms | 0 - 41 MB | NPU
| hand_detector | ONNX | float | Snapdragon® 8 Elite Mobile | 0.498 ms | 0 - 41 MB | NPU
| hand_detector | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.432 ms | 0 - 48 MB | NPU
| hand_detector | QNN_DLC | float | Snapdragon® X2 Elite | 0.606 ms | 1 - 1 MB | NPU
| hand_detector | QNN_DLC | float | Snapdragon® X Elite | 0.888 ms | 1 - 1 MB | NPU
| hand_detector | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 0.535 ms | 0 - 57 MB | NPU
| hand_detector | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 1.374 ms | 0 - 56 MB | NPU
| hand_detector | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 1.214 ms | 1 - 4 MB | NPU
| hand_detector | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 0.729 ms | 1 - 2 MB | NPU
| hand_detector | QNN_DLC | float | Qualcomm® SA8775P | 1.285 ms | 0 - 41 MB | NPU
| hand_detector | QNN_DLC | float | Qualcomm® SA8650P | 1.285 ms | 0 - 41 MB | NPU
| hand_detector | QNN_DLC | float | Qualcomm® SA8255P | 1.285 ms | 0 - 41 MB | NPU
| hand_detector | QNN_DLC | float | Qualcomm® QCS8450 | 1.374 ms | 0 - 56 MB | NPU
| hand_detector | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 1.111 ms | 3 - 5 MB | NPU
| hand_detector | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 0.888 ms | 1 - 1 MB | NPU
| hand_detector | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 0.44 ms | 1 - 39 MB | NPU
| hand_detector | QNN_DLC | float | Qualcomm® SA7255P | 3.776 ms | 1 - 39 MB | NPU
| hand_detector | QNN_DLC | float | Qualcomm® SA8295P | 1.692 ms | 0 - 31 MB | NPU
| hand_detector | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 0.44 ms | 1 - 39 MB | NPU
| hand_detector | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.388 ms | 0 - 40 MB | NPU
| hand_detector | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 0.541 ms | 0 - 55 MB | NPU
| hand_detector | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 1.374 ms | 0 - 56 MB | NPU
| hand_detector | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 1.219 ms | 0 - 8 MB | NPU
| hand_detector | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 0.732 ms | 0 - 2 MB | NPU
| hand_detector | TFLITE | float | Qualcomm® SA8775P | 1.302 ms | 0 - 41 MB | NPU
| hand_detector | TFLITE | float | Qualcomm® SA8650P | 1.302 ms | 0 - 41 MB | NPU
| hand_detector | TFLITE | float | Qualcomm® SA8255P | 1.302 ms | 0 - 41 MB | NPU
| hand_detector | TFLITE | float | Qualcomm® QCS8450 | 1.374 ms | 0 - 56 MB | NPU
| hand_detector | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 1.13 ms | 0 - 7 MB | NPU
| hand_detector | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 0.446 ms | 0 - 35 MB | NPU
| hand_detector | TFLITE | float | Qualcomm® SA7255P | 3.845 ms | 0 - 38 MB | NPU
| hand_detector | TFLITE | float | Qualcomm® SA8295P | 1.715 ms | 0 - 29 MB | NPU
| hand_detector | TFLITE | float | Snapdragon® 8 Elite Mobile | 0.446 ms | 0 - 35 MB | NPU
| hand_detector | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.391 ms | 0 - 36 MB | NPU
| hand_landmark_detector | ONNX | float | Snapdragon® X2 Elite | 0.713 ms | 2 - 2 MB | NPU
| hand_landmark_detector | ONNX | float | Snapdragon® X Elite | 1.145 ms | 6 - 6 MB | NPU
| hand_landmark_detector | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 0.816 ms | 0 - 67 MB | NPU
| hand_landmark_detector | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 2.089 ms | 1 - 60 MB | NPU
| hand_landmark_detector | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 2.008 ms | 1 - 4 MB | NPU
| hand_landmark_detector | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.182 ms | 0 - 37 MB | NPU
| hand_landmark_detector | ONNX | float | Qualcomm® QCS8450 | 2.089 ms | 1 - 60 MB | NPU
| hand_landmark_detector | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 1.839 ms | 1 - 4 MB | NPU
| hand_landmark_detector | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 1.145 ms | 6 - 6 MB | NPU
| hand_landmark_detector | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 0.659 ms | 0 - 38 MB | NPU
| hand_landmark_detector | ONNX | float | Snapdragon® 8 Elite Mobile | 0.659 ms | 0 - 38 MB | NPU
| hand_landmark_detector | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.571 ms | 0 - 44 MB | NPU
| hand_landmark_detector | QNN_DLC | float | Snapdragon® X2 Elite | 0.868 ms | 1 - 1 MB | NPU
| hand_landmark_detector | QNN_DLC | float | Snapdragon® X Elite | 1.292 ms | 1 - 1 MB | NPU
| hand_landmark_detector | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 0.765 ms | 0 - 56 MB | NPU
| hand_landmark_detector | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 1.932 ms | 0 - 55 MB | NPU
| hand_landmark_detector | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 1.833 ms | 1 - 4 MB | NPU
| hand_landmark_detector | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 0.999 ms | 1 - 2 MB | NPU
| hand_landmark_detector | QNN_DLC | float | Qualcomm® SA8775P | 1.878 ms | 0 - 37 MB | NPU
| hand_landmark_detector | QNN_DLC | float | Qualcomm® SA8650P | 1.878 ms | 0 - 37 MB | NPU
| hand_landmark_detector | QNN_DLC | float | Qualcomm® SA8255P | 1.878 ms | 0 - 37 MB | NPU
| hand_landmark_detector | QNN_DLC | float | Qualcomm® QCS8450 | 1.932 ms | 0 - 55 MB | NPU
| hand_landmark_detector | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 1.691 ms | 1 - 3 MB | NPU
| hand_landmark_detector | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 1.292 ms | 1 - 1 MB | NPU
| hand_landmark_detector | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 0.601 ms | 0 - 34 MB | NPU
| hand_landmark_detector | QNN_DLC | float | Qualcomm® SA7255P | 5.336 ms | 1 - 35 MB | NPU
| hand_landmark_detector | QNN_DLC | float | Qualcomm® SA8295P | 2.235 ms | 0 - 30 MB | NPU
| hand_landmark_detector | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 0.601 ms | 0 - 34 MB | NPU
| hand_landmark_detector | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.516 ms | 0 - 34 MB | NPU
| hand_landmark_detector | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 0.769 ms | 33 - 91 MB | NPU
| hand_landmark_detector | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 1.952 ms | 0 - 59 MB | NPU
| hand_landmark_detector | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 1.842 ms | 0 - 10 MB | NPU
| hand_landmark_detector | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 0.998 ms | 0 - 3 MB | NPU
| hand_landmark_detector | TFLITE | float | Qualcomm® SA8775P | 1.871 ms | 0 - 43 MB | NPU
| hand_landmark_detector | TFLITE | float | Qualcomm® SA8650P | 1.871 ms | 0 - 43 MB | NPU
| hand_landmark_detector | TFLITE | float | Qualcomm® SA8255P | 1.871 ms | 0 - 43 MB | NPU
| hand_landmark_detector | TFLITE | float | Qualcomm® QCS8450 | 1.952 ms | 0 - 59 MB | NPU
| hand_landmark_detector | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 1.683 ms | 0 - 9 MB | NPU
| hand_landmark_detector | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 0.608 ms | 0 - 37 MB | NPU
| hand_landmark_detector | TFLITE | float | Qualcomm® SA7255P | 9.224 ms | 0 - 30 MB | GPU
| hand_landmark_detector | TFLITE | float | Qualcomm® SA8295P | 2.221 ms | 0 - 35 MB | NPU
| hand_landmark_detector | TFLITE | float | Snapdragon® 8 Elite Mobile | 0.608 ms | 0 - 37 MB | NPU
| hand_landmark_detector | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.527 ms | 0 - 46 MB | NPU

## License
* The license for the original implementation of MediaPipe-Hand-Detection can be found
  [here](https://github.com/zmurez/MediaPipePyTorch/blob/master/LICENSE).

## References
* [MediaPipe Hands: On-device Real-time Hand Tracking](https://arxiv.org/abs/2006.10214)
* [Source Model Implementation](https://github.com/zmurez/MediaPipePyTorch/)

## Community
* For questions or feedback, chat with our engineers in [Qualcomm® AI Hub Workbench](https://workbench.aihub.qualcomm.cn).