---
title: PidNet
canonical_url: "https://www.modelscope.cn/models/Qualcomm-AI-Hub/PidNet"
md_url: "https://www.modelscope.cn/models/Qualcomm-AI-Hub/PidNet.md"
repository: Qualcomm-AI-Hub/PidNet
last_updated: 2026-10-16
license: mit
pipeline_tag: image-segmentation
tasks:
  - image-segmentation
downloads: 0
stars: 0
tags:
  - real_time
  - android
---

# PidNet

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

Qualcomm-AI-Hub/PidNet 是 ModelScope 魔搭社区上的image-segmentation模型，采用 mit 许可。

- **Repository**: Qualcomm-AI-Hub/PidNet
- **License**: mit
- **Tasks**: image-segmentation
- **Tags**: real_time, android
- **Downloads**: 0
- **Stars**: 0
- **Last updated**: 2026-10-16

Source: https://www.modelscope.cn/models/Qualcomm-AI-Hub/PidNet

---

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

# PidNet: Optimized for Qualcomm Devices

PIDNet (Proportional-Integral-Derivative Network) is a real-time semantic segmentation model based on  PID controllers

This is based on the implementation of PidNet found [here](https://github.com/XuJiacong/PIDNet).
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/pidnet) 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/pidnet/releases/v0.62.1/pidnet-onnx-float.zip)
| ONNX | w8a8 | 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/pidnet/releases/v0.62.1/pidnet-onnx-w8a8.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/pidnet/releases/v0.62.1/pidnet-qnn_dlc-float.zip)
| QNN_DLC | w8a8 | Universal | QAIRT 2.45 | [Download](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/pidnet/releases/v0.62.1/pidnet-qnn_dlc-w8a8.zip)
| TFLITE | float | Universal | QAIRT 2.45 | [Download](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/pidnet/releases/v0.62.1/pidnet-tflite-float.zip)
| TFLITE | w8a8 | Universal | QAIRT 2.45 | [Download](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/pidnet/releases/v0.62.1/pidnet-tflite-w8a8.zip)

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


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

## Model Details

**Model Type:** Model_use_case.semantic_segmentation

**Model Stats:**
- Inference latency: RealTime
- Input resolution: 1024x2048
- Model checkpoint: PIDNet_S_Cityscapes_val.pt
- Model size (float): 29.1 MB
- Model size (w8a8): 8.02 MB
- Number of output classes: 19
- Number of parameters: 8.06M

## Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|---|---|---|---|---|---|---
| PidNet | ONNX | float | Snapdragon® X2 Elite | 13.743 ms | 24 - 24 MB | NPU
| PidNet | ONNX | float | Snapdragon® X Elite | 32.114 ms | 24 - 24 MB | NPU
| PidNet | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 22.848 ms | 30 - 331 MB | NPU
| PidNet | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 54.789 ms | 13 - 318 MB | NPU
| PidNet | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 40.42 ms | 24 - 51 MB | NPU
| PidNet | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 32.832 ms | 24 - 46 MB | NPU
| PidNet | ONNX | float | Qualcomm® QCS8450 | 54.789 ms | 13 - 318 MB | NPU
| PidNet | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 42.261 ms | 24 - 51 MB | NPU
| PidNet | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 32.114 ms | 24 - 24 MB | NPU
| PidNet | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 17.614 ms | 6 - 219 MB | NPU
| PidNet | ONNX | float | Snapdragon® 8 Elite Mobile | 17.614 ms | 6 - 219 MB | NPU
| PidNet | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 13.235 ms | 7 - 256 MB | NPU
| PidNet | ONNX | w8a8 | Snapdragon® X2 Elite | 26.075 ms | 7 - 7 MB | NPU
| PidNet | ONNX | w8a8 | Snapdragon® X Elite | 54.893 ms | 7 - 7 MB | NPU
| PidNet | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 38.835 ms | 7 - 270 MB | NPU
| PidNet | ONNX | w8a8 | Snapdragon® 8 Gen 1 Mobile | 57.517 ms | 8 - 271 MB | NPU
| PidNet | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 157.864 ms | 6 - 9 MB | NPU
| PidNet | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 51.006 ms | 6 - 10 MB | NPU
| PidNet | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 52.717 ms | 6 - 177 MB | NPU
| PidNet | ONNX | w8a8 | Qualcomm® QCS8450 | 57.517 ms | 8 - 271 MB | NPU
| PidNet | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 54.064 ms | 6 - 9 MB | NPU
| PidNet | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 54.893 ms | 7 - 7 MB | NPU
| PidNet | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 210.332 ms | 6 - 225 MB | NPU
| PidNet | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 62.654 ms | 7 - 235 MB | NPU
| PidNet | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 44.899 ms | 1 - 217 MB | NPU
| PidNet | ONNX | w8a8 | Snapdragon® 8 Elite Mobile | 44.899 ms | 1 - 217 MB | NPU
| PidNet | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 25.8 ms | 2 - 236 MB | NPU
| PidNet | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 62.654 ms | 7 - 235 MB | NPU
| PidNet | QNN_DLC | float | Snapdragon® X2 Elite | 13.562 ms | 24 - 24 MB | NPU
| PidNet | QNN_DLC | float | Snapdragon® X Elite | 38.875 ms | 24 - 24 MB | NPU
| PidNet | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 25.468 ms | 16 - 323 MB | NPU
| PidNet | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 75.957 ms | 3 - 308 MB | NPU
| PidNet | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 44.511 ms | 24 - 53 MB | NPU
| PidNet | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 37.729 ms | 24 - 27 MB | NPU
| PidNet | QNN_DLC | float | Qualcomm® SA8775P | 46.605 ms | 24 - 247 MB | NPU
| PidNet | QNN_DLC | float | Qualcomm® SA8650P | 46.605 ms | 24 - 247 MB | NPU
| PidNet | QNN_DLC | float | Qualcomm® SA8255P | 46.605 ms | 24 - 247 MB | NPU
| PidNet | QNN_DLC | float | Qualcomm® QCS8450 | 75.957 ms | 3 - 308 MB | NPU
| PidNet | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 47.528 ms | 24 - 52 MB | NPU
| PidNet | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 38.875 ms | 24 - 24 MB | NPU
| PidNet | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 18.213 ms | 23 - 267 MB | NPU
| PidNet | QNN_DLC | float | Qualcomm® SA7255P | 117.751 ms | 24 - 246 MB | NPU
| PidNet | QNN_DLC | float | Qualcomm® SA8295P | 51.907 ms | 24 - 252 MB | NPU
| PidNet | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 18.213 ms | 23 - 267 MB | NPU
| PidNet | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 12.062 ms | 18 - 289 MB | NPU
| PidNet | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 25.762 ms | 6 - 6 MB | NPU
| PidNet | QNN_DLC | w8a8 | Snapdragon® X Elite | 60.63 ms | 6 - 6 MB | NPU
| PidNet | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 43.111 ms | 6 - 267 MB | NPU
| PidNet | QNN_DLC | w8a8 | Snapdragon® 8 Gen 1 Mobile | 63.574 ms | 6 - 270 MB | NPU
| PidNet | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 56.547 ms | 6 - 15 MB | NPU
| PidNet | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 57.901 ms | 6 - 8 MB | NPU
| PidNet | QNN_DLC | w8a8 | Qualcomm® SA8775P | 58.537 ms | 6 - 219 MB | NPU
| PidNet | QNN_DLC | w8a8 | Qualcomm® SA8650P | 58.537 ms | 6 - 219 MB | NPU
| PidNet | QNN_DLC | w8a8 | Qualcomm® SA8255P | 58.537 ms | 6 - 219 MB | NPU
| PidNet | QNN_DLC | w8a8 | Qualcomm® QCS8450 | 63.574 ms | 6 - 270 MB | NPU
| PidNet | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 59.064 ms | 6 - 14 MB | NPU
| PidNet | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 60.63 ms | 6 - 6 MB | NPU
| PidNet | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 43.833 ms | 6 - 241 MB | NPU
| PidNet | QNN_DLC | w8a8 | Qualcomm® SA7255P | 111.291 ms | 6 - 218 MB | NPU
| PidNet | QNN_DLC | w8a8 | Qualcomm® SA8295P | 66.453 ms | 6 - 223 MB | NPU
| PidNet | QNN_DLC | w8a8 | Snapdragon® 8 Elite Mobile | 43.833 ms | 6 - 241 MB | NPU
| PidNet | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 24.54 ms | 6 - 272 MB | NPU
| PidNet | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 24.955 ms | 2 - 322 MB | NPU
| PidNet | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 74.253 ms | 0 - 326 MB | NPU
| PidNet | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 43.808 ms | 0 - 46 MB | NPU
| PidNet | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 36.276 ms | 2 - 5 MB | NPU
| PidNet | TFLITE | float | Qualcomm® SA8775P | 45.283 ms | 3 - 233 MB | NPU
| PidNet | TFLITE | float | Qualcomm® SA8650P | 45.283 ms | 3 - 233 MB | NPU
| PidNet | TFLITE | float | Qualcomm® SA8255P | 45.283 ms | 3 - 233 MB | NPU
| PidNet | TFLITE | float | Qualcomm® QCS8450 | 74.253 ms | 0 - 326 MB | NPU
| PidNet | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 45.833 ms | 0 - 45 MB | NPU
| PidNet | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 18.901 ms | 2 - 255 MB | NPU
| PidNet | TFLITE | float | Qualcomm® SA7255P | 116.628 ms | 0 - 228 MB | NPU
| PidNet | TFLITE | float | Qualcomm® SA8295P | 50.971 ms | 3 - 245 MB | NPU
| PidNet | TFLITE | float | Snapdragon® 8 Elite Mobile | 18.901 ms | 2 - 255 MB | NPU
| PidNet | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 12.648 ms | 4 - 274 MB | NPU
| PidNet | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 38.123 ms | 1 - 262 MB | NPU
| PidNet | TFLITE | w8a8 | Snapdragon® 8 Gen 1 Mobile | 58.254 ms | 1 - 266 MB | NPU
| PidNet | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 200.632 ms | 2 - 72 MB | NPU
| PidNet | TFLITE | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 49.767 ms | 1 - 18 MB | NPU
| PidNet | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 50.412 ms | 0 - 3 MB | NPU
| PidNet | TFLITE | w8a8 | Qualcomm® SA8775P | 51.017 ms | 1 - 214 MB | NPU
| PidNet | TFLITE | w8a8 | Qualcomm® SA8650P | 51.017 ms | 1 - 214 MB | NPU
| PidNet | TFLITE | w8a8 | Qualcomm® SA8255P | 51.017 ms | 1 - 214 MB | NPU
| PidNet | TFLITE | w8a8 | Qualcomm® QCS8450 | 58.254 ms | 1 - 266 MB | NPU
| PidNet | TFLITE | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 51.628 ms | 0 - 17 MB | NPU
| PidNet | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 243.409 ms | 3 - 233 MB | NPU
| PidNet | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 66.153 ms | 1 - 219 MB | NPU
| PidNet | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 70.041 ms | 1 - 233 MB | NPU
| PidNet | TFLITE | w8a8 | Qualcomm® SA7255P | 98.332 ms | 0 - 212 MB | NPU
| PidNet | TFLITE | w8a8 | Qualcomm® SA8295P | 58.069 ms | 1 - 217 MB | NPU
| PidNet | TFLITE | w8a8 | Snapdragon® 8 Elite Mobile | 70.041 ms | 1 - 233 MB | NPU
| PidNet | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 22.173 ms | 0 - 266 MB | NPU
| PidNet | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 66.153 ms | 1 - 219 MB | NPU

## License
* The license for the original implementation of PidNet can be found
  [here](https://github.com/XuJiacong/PIDNet/blob/main/LICENSE).

## References
* [PIDNet A Real-time Semantic Segmentation Network Inspired from PID Controller Segmentation of Road Scenes](https://arxiv.org/abs/2206.02066)
* [Source Model Implementation](https://github.com/XuJiacong/PIDNet)

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