README.md

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

---

![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/web-assets/model_demo.png)

# DeepLabV3-Plus-MobileNet: Optimized for Qualcomm Devices

DeepLabV3 is designed for semantic segmentation at multiple scales, trained on the various datasets. It uses MobileNet as a backbone.

This is based on the implementation of DeepLabV3-Plus-MobileNet found [here](https://github.com/jfzhang95/pytorch-deeplab-xception).
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/v0.59.0/src/qai_hub_models/models/deeplabv3_plus_mobilenet) 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 |
|---|---|---|---|---|
| ONNX | float | Universal | QAIRT 2.45, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/releases/v0.59.0/deeplabv3_plus_mobilenet-onnx-float.zip)
| ONNX | w8a16 | Universal | QAIRT 2.45, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/releases/v0.59.0/deeplabv3_plus_mobilenet-onnx-w8a16.zip)
| ONNX | w8a8 | Universal | QAIRT 2.45, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/releases/v0.59.0/deeplabv3_plus_mobilenet-onnx-w8a8.zip)
| QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/releases/v0.59.0/deeplabv3_plus_mobilenet-qnn_dlc-float.zip)
| QNN_DLC | w8a16 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/releases/v0.59.0/deeplabv3_plus_mobilenet-qnn_dlc-w8a16.zip)
| QNN_DLC | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/releases/v0.59.0/deeplabv3_plus_mobilenet-qnn_dlc-w8a8.zip)
| TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/releases/v0.59.0/deeplabv3_plus_mobilenet-tflite-float.zip)
| TFLITE | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/releases/v0.59.0/deeplabv3_plus_mobilenet-tflite-w8a8.zip)

For more device-specific assets and performance metrics, visit **[DeepLabV3-Plus-MobileNet on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/deeplabv3_plus_mobilenet)**.


### Option 2: Export with Custom Configurations

Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.59.0/src/qai_hub_models/models/deeplabv3_plus_mobilenet) 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 [DeepLabV3-Plus-MobileNet on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.59.0/src/qai_hub_models/models/deeplabv3_plus_mobilenet) for usage instructions.

## Model Details

**Model Type:** Model_use_case.semantic_segmentation

**Model Stats:**
- Model checkpoint: VOC2012
- Input resolution: 513x513
- Number of output classes: 21
- Number of parameters: 5.80M
- Model size (float): 22.2 MB
- Model size (w8a16): 6.67 MB

## Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|---|---|---|---|---|---|---
| DeepLabV3-Plus-MobileNet | ONNX | float | Snapdragon® X2 Elite | 5.211 ms | 3 - 3 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | float | Snapdragon® X Elite | 10.876 ms | 11 - 11 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 6.981 ms | 4 - 204 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 18.03 ms | 4 - 209 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 10.285 ms | 0 - 13 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | float | Qualcomm® QCS8450 | 18.03 ms | 4 - 209 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 16.619 ms | 3 - 6 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 10.876 ms | 11 - 11 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 5.986 ms | 1 - 170 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | float | Snapdragon® 8 Elite Mobile | 5.986 ms | 1 - 170 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 4.352 ms | 1 - 174 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Snapdragon® X2 Elite | 3.448 ms | 2 - 2 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Snapdragon® X Elite | 7.343 ms | 5 - 5 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 5.109 ms | 2 - 215 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Snapdragon® 8 Gen 1 Mobile | 9.575 ms | 0 - 215 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS6490 | 35.798 ms | 2 - 4 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 6.995 ms | 0 - 8 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Qualcomm® QCS8450 | 9.575 ms | 0 - 215 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 7.166 ms | 2 - 4 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 7.343 ms | 5 - 5 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 3.875 ms | 0 - 188 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Snapdragon® 8 Elite Mobile | 3.875 ms | 0 - 188 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 2.782 ms | 0 - 187 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Snapdragon® X2 Elite | 1.587 ms | 2 - 2 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Snapdragon® X Elite | 3.654 ms | 6 - 6 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 2.403 ms | 0 - 201 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Snapdragon® 8 Gen 1 Mobile | 4.818 ms | 1 - 205 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 16.427 ms | 1 - 4 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 3.465 ms | 0 - 9 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Qualcomm® QCS8450 | 4.818 ms | 1 - 205 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 3.847 ms | 1 - 4 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 3.654 ms | 6 - 6 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 1.906 ms | 0 - 180 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Snapdragon® 8 Elite Mobile | 1.906 ms | 0 - 180 MB | NPU
| DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 1.413 ms | 0 - 182 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | float | Snapdragon® X2 Elite | 5.55 ms | 3 - 3 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | float | Snapdragon® X Elite | 12.316 ms | 3 - 3 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 8.078 ms | 0 - 208 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 19.799 ms | 3 - 208 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8275 | 58.593 ms | 1 - 166 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 11.575 ms | 3 - 5 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® SA8775P | 17.495 ms | 1 - 167 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® SA8650P | 17.495 ms | 1 - 167 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® SA8255P | 17.495 ms | 1 - 167 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® QCS8450 | 19.799 ms | 3 - 208 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 17.872 ms | 3 - 8 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 12.316 ms | 3 - 3 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 6.276 ms | 0 - 177 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® SA7255P | 58.593 ms | 1 - 166 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® SA8295P | 19.577 ms | 0 - 169 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 6.276 ms | 0 - 177 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 4.385 ms | 3 - 183 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 4.247 ms | 2 - 2 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Snapdragon® X Elite | 9.199 ms | 2 - 2 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 6.122 ms | 0 - 213 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Snapdragon® 8 Gen 1 Mobile | 12.117 ms | 2 - 215 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ QCS6490 | 39.642 ms | 3 - 6 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ QCS8275 | 21.986 ms | 2 - 182 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 8.396 ms | 2 - 3 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® SA8775P | 8.911 ms | 2 - 182 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® SA8650P | 8.911 ms | 2 - 182 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® SA8255P | 8.911 ms | 2 - 182 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® QCS8450 | 12.117 ms | 2 - 215 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 8.637 ms | 2 - 5 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 9.199 ms | 2 - 2 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 110.006 ms | 2 - 232 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 12.054 ms | 2 - 180 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 4.346 ms | 2 - 182 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® SA7255P | 21.986 ms | 2 - 182 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® SA8295P | 13.756 ms | 2 - 197 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Snapdragon® 8 Elite Mobile | 4.346 ms | 2 - 182 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 3.134 ms | 2 - 196 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 12.054 ms | 2 - 180 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 2.093 ms | 1 - 1 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Snapdragon® X Elite | 4.618 ms | 1 - 1 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 2.917 ms | 1 - 194 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Snapdragon® 8 Gen 1 Mobile | 6.659 ms | 1 - 197 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 17.861 ms | 3 - 5 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS8275 | 10.82 ms | 1 - 169 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 4.163 ms | 1 - 3 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® SA8775P | 4.71 ms | 1 - 171 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® SA8650P | 4.71 ms | 1 - 171 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® SA8255P | 4.71 ms | 1 - 171 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® QCS8450 | 6.659 ms | 1 - 197 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 4.46 ms | 0 - 3 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 4.618 ms | 1 - 1 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 50.731 ms | 1 - 201 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 6.284 ms | 1 - 176 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 2.187 ms | 1 - 173 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® SA7255P | 10.82 ms | 1 - 169 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® SA8295P | 6.415 ms | 1 - 168 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Snapdragon® 8 Elite Mobile | 2.187 ms | 1 - 173 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 1.639 ms | 1 - 178 MB | NPU
| DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 6.284 ms | 1 - 176 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 8.08 ms | 0 - 213 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 19.844 ms | 0 - 212 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | float | Qualcomm® Dragonwing™ QCS8275 | 58.675 ms | 0 - 168 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 11.523 ms | 0 - 2 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | float | Qualcomm® SA8775P | 17.546 ms | 0 - 173 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | float | Qualcomm® SA8650P | 17.546 ms | 0 - 173 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | float | Qualcomm® SA8255P | 17.546 ms | 0 - 173 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | float | Qualcomm® QCS8450 | 19.844 ms | 0 - 212 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 17.352 ms | 0 - 18 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 6.28 ms | 0 - 177 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | float | Qualcomm® SA7255P | 58.675 ms | 0 - 168 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | float | Qualcomm® SA8295P | 19.614 ms | 0 - 169 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | float | Snapdragon® 8 Elite Mobile | 6.28 ms | 0 - 177 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 4.393 ms | 0 - 183 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 2.944 ms | 0 - 202 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Snapdragon® 8 Gen 1 Mobile | 5.789 ms | 0 - 203 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 18.845 ms | 0 - 11 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS8275 | 10.834 ms | 0 - 174 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 4.329 ms | 0 - 2 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® SA8775P | 4.619 ms | 0 - 177 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® SA8650P | 4.619 ms | 0 - 177 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® SA8255P | 4.619 ms | 0 - 177 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® QCS8450 | 5.789 ms | 0 - 203 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 4.371 ms | 0 - 9 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 52.286 ms | 0 - 196 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 6.32 ms | 0 - 177 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 2.08 ms | 0 - 177 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® SA7255P | 10.834 ms | 0 - 174 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® SA8295P | 6.228 ms | 0 - 172 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Snapdragon® 8 Elite Mobile | 2.08 ms | 0 - 177 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 1.609 ms | 0 - 184 MB | NPU
| DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 6.32 ms | 0 - 177 MB | NPU

## License
* The license for the original implementation of DeepLabV3-Plus-MobileNet can be found
  [here](https://github.com/jfzhang95/pytorch-deeplab-xception/blob/master/LICENSE).

## References
* [Rethinking Atrous Convolution for Semantic Image Segmentation](https://arxiv.org/abs/1706.05587)
* [Source Model Implementation](https://github.com/jfzhang95/pytorch-deeplab-xception)

## 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).