DeepLabXception: Optimized for Mobile Deployment

Deep Convolutional Neural Network model for semantic segmentation

DeepLabXception is a semantic segmentation model supporting multiple backbones like ResNet-101 and Xception, with flexible dataset compatibility including COCO, VOC, and Cityscapes.

This model is an implementation of DeepLabXception found here.

This repository provides scripts to run DeepLabXception on Qualcomm® devices. More details on model performance across various devices, can be found here.

Model Details

Model Type: Model_use_case.semantic_segmentation
Model Stats:
- Model checkpoint: COCO_WITH_VOC_LABELS_V1
- Input resolution: 480x520
- Number of output classes: 21
- Number of parameters: 41.26M
- Model size (float): 158 MB

Model	Precision	Device	Chipset	Target Runtime	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit	Target Model
DeepLabXception	float	QCS8275 (Proxy)	Qualcomm® QCS8275 (Proxy)	TFLITE	113.083 ms	0 - 317 MB	NPU	DeepLabXception.tflite
DeepLabXception	float	QCS8275 (Proxy)	Qualcomm® QCS8275 (Proxy)	QNN_DLC	113.025 ms	3 - 212 MB	NPU	DeepLabXception.dlc
DeepLabXception	float	QCS8450 (Proxy)	Qualcomm® QCS8450 (Proxy)	TFLITE	46.39 ms	0 - 434 MB	NPU	DeepLabXception.tflite
DeepLabXception	float	QCS8450 (Proxy)	Qualcomm® QCS8450 (Proxy)	QNN_DLC	50.98 ms	3 - 329 MB	NPU	DeepLabXception.dlc
DeepLabXception	float	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	TFLITE	21.884 ms	0 - 3 MB	NPU	DeepLabXception.tflite
DeepLabXception	float	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	QNN_DLC	21.772 ms	3 - 5 MB	NPU	DeepLabXception.dlc
DeepLabXception	float	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	ONNX	21.042 ms	0 - 96 MB	NPU	DeepLabXception.onnx.zip
DeepLabXception	float	QCS9075 (Proxy)	Qualcomm® QCS9075 (Proxy)	TFLITE	30.499 ms	0 - 319 MB	NPU	DeepLabXception.tflite
DeepLabXception	float	QCS9075 (Proxy)	Qualcomm® QCS9075 (Proxy)	QNN_DLC	30.466 ms	0 - 209 MB	NPU	DeepLabXception.dlc
DeepLabXception	float	SA7255P ADP	Qualcomm® SA7255P	TFLITE	113.083 ms	0 - 317 MB	NPU	DeepLabXception.tflite
DeepLabXception	float	SA7255P ADP	Qualcomm® SA7255P	QNN_DLC	113.025 ms	3 - 212 MB	NPU	DeepLabXception.dlc
DeepLabXception	float	SA8295P ADP	Qualcomm® SA8295P	TFLITE	38.81 ms	0 - 315 MB	NPU	DeepLabXception.tflite
DeepLabXception	float	SA8295P ADP	Qualcomm® SA8295P	QNN_DLC	38.868 ms	0 - 213 MB	NPU	DeepLabXception.dlc
DeepLabXception	float	SA8775P ADP	Qualcomm® SA8775P	TFLITE	30.499 ms	0 - 319 MB	NPU	DeepLabXception.tflite
DeepLabXception	float	SA8775P ADP	Qualcomm® SA8775P	QNN_DLC	30.466 ms	0 - 209 MB	NPU	DeepLabXception.dlc
DeepLabXception	float	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	TFLITE	15.932 ms	0 - 440 MB	NPU	DeepLabXception.tflite
DeepLabXception	float	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	QNN_DLC	15.952 ms	3 - 334 MB	NPU	DeepLabXception.dlc
DeepLabXception	float	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	ONNX	15.679 ms	3 - 304 MB	NPU	DeepLabXception.onnx.zip
DeepLabXception	float	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	TFLITE	12.31 ms	0 - 319 MB	NPU	DeepLabXception.tflite
DeepLabXception	float	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	QNN_DLC	12.284 ms	3 - 215 MB	NPU	DeepLabXception.dlc
DeepLabXception	float	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	ONNX	12.273 ms	3 - 184 MB	NPU	DeepLabXception.onnx.zip
DeepLabXception	float	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen 5 Mobile	TFLITE	9.179 ms	0 - 328 MB	NPU	DeepLabXception.tflite
DeepLabXception	float	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen 5 Mobile	QNN_DLC	9.226 ms	3 - 236 MB	NPU	DeepLabXception.dlc
DeepLabXception	float	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen 5 Mobile	ONNX	9.24 ms	3 - 200 MB	NPU	DeepLabXception.onnx.zip
DeepLabXception	float	Snapdragon X Elite CRD	Snapdragon® X Elite	QNN_DLC	22.62 ms	3 - 3 MB	NPU	DeepLabXception.dlc
DeepLabXception	float	Snapdragon X Elite CRD	Snapdragon® X Elite	ONNX	21.775 ms	85 - 85 MB	NPU	DeepLabXception.onnx.zip
DeepLabXception	w8a8	Dragonwing Q-6690 MTP	Qualcomm® QCM6690	TFLITE	89.688 ms	0 - 270 MB	NPU	DeepLabXception.tflite
DeepLabXception	w8a8	Dragonwing Q-6690 MTP	Qualcomm® QCM6690	QNN_DLC	112.359 ms	1 - 278 MB	NPU	DeepLabXception.dlc
DeepLabXception	w8a8	Dragonwing Q-6690 MTP	Qualcomm® QCM6690	ONNX	240.789 ms	88 - 106 MB	CPU	DeepLabXception.onnx.zip
DeepLabXception	w8a8	Dragonwing RB3 Gen 2 Vision Kit	Qualcomm® QCS6490	TFLITE	28.818 ms	0 - 58 MB	NPU	DeepLabXception.tflite
DeepLabXception	w8a8	Dragonwing RB3 Gen 2 Vision Kit	Qualcomm® QCS6490	QNN_DLC	30.262 ms	1 - 3 MB	NPU	DeepLabXception.dlc
DeepLabXception	w8a8	Dragonwing RB3 Gen 2 Vision Kit	Qualcomm® QCS6490	ONNX	310.37 ms	90 - 118 MB	CPU	DeepLabXception.onnx.zip
DeepLabXception	w8a8	QCS8275 (Proxy)	Qualcomm® QCS8275 (Proxy)	TFLITE	18.973 ms	0 - 257 MB	NPU	DeepLabXception.tflite
DeepLabXception	w8a8	QCS8275 (Proxy)	Qualcomm® QCS8275 (Proxy)	QNN_DLC	20.382 ms	1 - 261 MB	NPU	DeepLabXception.dlc
DeepLabXception	w8a8	QCS8450 (Proxy)	Qualcomm® QCS8450 (Proxy)	TFLITE	11.81 ms	0 - 338 MB	NPU	DeepLabXception.tflite
DeepLabXception	w8a8	QCS8450 (Proxy)	Qualcomm® QCS8450 (Proxy)	QNN_DLC	14.937 ms	1 - 342 MB	NPU	DeepLabXception.dlc
DeepLabXception	w8a8	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	TFLITE	6.783 ms	0 - 3 MB	NPU	DeepLabXception.tflite
DeepLabXception	w8a8	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	QNN_DLC	8.067 ms	1 - 3 MB	NPU	DeepLabXception.dlc
DeepLabXception	w8a8	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	ONNX	7.651 ms	0 - 51 MB	NPU	DeepLabXception.onnx.zip
DeepLabXception	w8a8	QCS9075 (Proxy)	Qualcomm® QCS9075 (Proxy)	TFLITE	7.358 ms	0 - 255 MB	NPU	DeepLabXception.tflite
DeepLabXception	w8a8	QCS9075 (Proxy)	Qualcomm® QCS9075 (Proxy)	QNN_DLC	8.13 ms	1 - 258 MB	NPU	DeepLabXception.dlc
DeepLabXception	w8a8	SA7255P ADP	Qualcomm® SA7255P	TFLITE	18.973 ms	0 - 257 MB	NPU	DeepLabXception.tflite
DeepLabXception	w8a8	SA7255P ADP	Qualcomm® SA7255P	QNN_DLC	20.382 ms	1 - 261 MB	NPU	DeepLabXception.dlc
DeepLabXception	w8a8	SA8295P ADP	Qualcomm® SA8295P	TFLITE	11.583 ms	0 - 245 MB	NPU	DeepLabXception.tflite
DeepLabXception	w8a8	SA8295P ADP	Qualcomm® SA8295P	QNN_DLC	12.597 ms	1 - 251 MB	NPU	DeepLabXception.dlc
DeepLabXception	w8a8	SA8775P ADP	Qualcomm® SA8775P	TFLITE	7.358 ms	0 - 255 MB	NPU	DeepLabXception.tflite
DeepLabXception	w8a8	SA8775P ADP	Qualcomm® SA8775P	QNN_DLC	8.13 ms	1 - 258 MB	NPU	DeepLabXception.dlc
DeepLabXception	w8a8	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	TFLITE	4.997 ms	0 - 343 MB	NPU	DeepLabXception.tflite
DeepLabXception	w8a8	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	QNN_DLC	5.562 ms	1 - 349 MB	NPU	DeepLabXception.dlc
DeepLabXception	w8a8	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	ONNX	5.295 ms	1 - 337 MB	NPU	DeepLabXception.onnx.zip
DeepLabXception	w8a8	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	TFLITE	3.918 ms	0 - 240 MB	NPU	DeepLabXception.tflite
DeepLabXception	w8a8	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	QNN_DLC	4.362 ms	1 - 259 MB	NPU	DeepLabXception.dlc
DeepLabXception	w8a8	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	ONNX	4.332 ms	1 - 230 MB	NPU	DeepLabXception.onnx.zip
DeepLabXception	w8a8	Snapdragon 7 Gen 4 QRD	Snapdragon® 7 Gen 4 Mobile	TFLITE	11.013 ms	0 - 242 MB	NPU	DeepLabXception.tflite
DeepLabXception	w8a8	Snapdragon 7 Gen 4 QRD	Snapdragon® 7 Gen 4 Mobile	QNN_DLC	12.18 ms	1 - 250 MB	NPU	DeepLabXception.dlc
DeepLabXception	w8a8	Snapdragon 7 Gen 4 QRD	Snapdragon® 7 Gen 4 Mobile	ONNX	227.149 ms	92 - 111 MB	CPU	DeepLabXception.onnx.zip
DeepLabXception	w8a8	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen 5 Mobile	TFLITE	2.886 ms	0 - 247 MB	NPU	DeepLabXception.tflite
DeepLabXception	w8a8	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen 5 Mobile	QNN_DLC	3.234 ms	1 - 256 MB	NPU	DeepLabXception.dlc
DeepLabXception	w8a8	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen 5 Mobile	ONNX	3.33 ms	0 - 235 MB	NPU	DeepLabXception.onnx.zip
DeepLabXception	w8a8	Snapdragon X Elite CRD	Snapdragon® X Elite	QNN_DLC	8.697 ms	1 - 1 MB	NPU	DeepLabXception.dlc
DeepLabXception	w8a8	Snapdragon X Elite CRD	Snapdragon® X Elite	ONNX	8.042 ms	44 - 44 MB	NPU	DeepLabXception.onnx.zip

Installation

Install the package via pip:

pip install qai-hub-models

Configure Qualcomm® AI Hub Workbench to run this model on a cloud-hosted device

Sign-in to Qualcomm® AI Hub Workbench with your Qualcomm® ID. Once signed in navigate to Account -> Settings -> API Token.

With this API token, you can configure your client to run models on the cloud hosted devices.

qai-hub configure --api_token API_TOKEN

Navigate to docs for more information.

Demo off target

The package contains a simple end-to-end demo that downloads pre-trained weights and runs this model on a sample input.

python -m qai_hub_models.models.deeplab_xception.demo

The above demo runs a reference implementation of pre-processing, model inference, and post processing.

NOTE: If you want running in a Jupyter Notebook or Google Colab like environment, please add the following to your cell (instead of the above).

%run -m qai_hub_models.models.deeplab_xception.demo

Run model on a cloud-hosted device

In addition to the demo, you can also run the model on a cloud-hosted Qualcomm® device. This script does the following:

Performance check on-device on a cloud-hosted device
Downloads compiled assets that can be deployed on-device for Android.
Accuracy check between PyTorch and on-device outputs.

python -m qai_hub_models.models.deeplab_xception.export

How does this work?

This export script leverages Qualcomm® AI Hub to optimize, validate, and deploy this model on-device. Lets go through each step below in detail:

Step 1: Compile model for on-device deployment

To compile a PyTorch model for on-device deployment, we first trace the model in memory using the jit.trace and then call the submit_compile_job API.

import torch

import qai_hub as hub
from qai_hub_models.models.deeplab_xception import Model

# Load the model
torch_model = Model.from_pretrained()

# Device
device = hub.Device("Samsung Galaxy S25")

# Trace model
input_shape = torch_model.get_input_spec()
sample_inputs = torch_model.sample_inputs()

pt_model = torch.jit.trace(torch_model, [torch.tensor(data[0]) for _, data in sample_inputs.items()])

# Compile model on a specific device
compile_job = hub.submit_compile_job(
    model=pt_model,
    device=device,
    input_specs=torch_model.get_input_spec(),
)

# Get target model to run on-device
target_model = compile_job.get_target_model()

Step 2: Performance profiling on cloud-hosted device

After compiling models from step 1. Models can be profiled model on-device using the target_model. Note that this scripts runs the model on a device automatically provisioned in the cloud. Once the job is submitted, you can navigate to a provided job URL to view a variety of on-device performance metrics.

profile_job = hub.submit_profile_job(
    model=target_model,
    device=device,
)

Step 3: Verify on-device accuracy

To verify the accuracy of the model on-device, you can run on-device inference on sample input data on the same cloud hosted device.

input_data = torch_model.sample_inputs()
inference_job = hub.submit_inference_job(
    model=target_model,
    device=device,
    inputs=input_data,
)
    on_device_output = inference_job.download_output_data()

With the output of the model, you can compute like PSNR, relative errors or spot check the output with expected output.

Note: This on-device profiling and inference requires access to Qualcomm® AI Hub Workbench. Sign up for access.

Run demo on a cloud-hosted device

You can also run the demo on-device.

python -m qai_hub_models.models.deeplab_xception.demo --eval-mode on-device

NOTE: If you want running in a Jupyter Notebook or Google Colab like environment, please add the following to your cell (instead of the above).

%run -m qai_hub_models.models.deeplab_xception.demo -- --eval-mode on-device

Deploying compiled model to Android

The models can be deployed using multiple runtimes:

TensorFlow Lite (.tflite export): This tutorial provides a guide to deploy the .tflite model in an Android application.
QNN (.so export ): This sample app provides instructions on how to use the .so shared library in an Android application.

View on Qualcomm® AI Hub

Get more details on DeepLabXception's performance across various devices here. Explore all available models on Qualcomm® AI Hub

License

The license for the original implementation of DeepLabXception can be found here.

References

Community

Join our AI Hub Slack community to collaborate, post questions and learn more about on-device AI.
For questions or feedback please reach out to us.

Downloads last month: 55

Paper for qualcomm/DeepLabXception

Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation

Paper • 1802.02611 • Published Feb 7, 2018