DeepLabXception: Optimized for Qualcomm Devices

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

This is based on the implementation of DeepLabXception found here. This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the Qualcomm® AI Hub Models library to export with custom configurations. More details on model performance across various devices, can be found here.

Qualcomm AI Hub Models uses Qualcomm AI Hub Workbench to compile, profile, and evaluate this model. Sign up 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 2.42, ONNX Runtime 1.24.1	Download
ONNX	w8a8	Universal	QAIRT 2.42, ONNX Runtime 1.24.1	Download
QNN_DLC	float	Universal	QAIRT 2.43	Download
QNN_DLC	w8a8	Universal	QAIRT 2.43	Download
TFLITE	float	Universal	QAIRT 2.43, TFLite 2.17.0	Download
TFLITE	w8a8	Universal	QAIRT 2.43, TFLite 2.17.0	Download

For more device-specific assets and performance metrics, visit DeepLabXception on Qualcomm® AI Hub.

Option 2: Export with Custom Configurations

Use the Qualcomm® AI Hub Models 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 DeepLabXception on GitHub for usage instructions.

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

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
DeepLabXception	ONNX	float	Snapdragon® X2 Elite	10.281 ms	86 - 86 MB	NPU
DeepLabXception	ONNX	float	Snapdragon® X Elite	22.717 ms	85 - 85 MB	NPU
DeepLabXception	ONNX	float	Snapdragon® 8 Gen 3 Mobile	16.878 ms	3 - 368 MB	NPU
DeepLabXception	ONNX	float	Qualcomm® QCS8550 (Proxy)	22.395 ms	0 - 96 MB	NPU
DeepLabXception	ONNX	float	Qualcomm® QCS9075	32.127 ms	3 - 8 MB	NPU
DeepLabXception	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	13.119 ms	1 - 252 MB	NPU
DeepLabXception	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	8.729 ms	3 - 255 MB	NPU
DeepLabXception	ONNX	w8a8	Snapdragon® X2 Elite	3.341 ms	44 - 44 MB	NPU
DeepLabXception	ONNX	w8a8	Snapdragon® X Elite	8.009 ms	44 - 44 MB	NPU
DeepLabXception	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	5.249 ms	0 - 388 MB	NPU
DeepLabXception	ONNX	w8a8	Qualcomm® QCS6490	311.091 ms	91 - 113 MB	CPU
DeepLabXception	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	7.436 ms	0 - 61 MB	NPU
DeepLabXception	ONNX	w8a8	Qualcomm® QCS9075	7.611 ms	0 - 4 MB	NPU
DeepLabXception	ONNX	w8a8	Qualcomm® QCM6690	244.195 ms	59 - 72 MB	CPU
DeepLabXception	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	4.239 ms	0 - 281 MB	NPU
DeepLabXception	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	229.357 ms	89 - 102 MB	CPU
DeepLabXception	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	3.06 ms	0 - 292 MB	NPU
DeepLabXception	QNN_DLC	float	Snapdragon® X2 Elite	11.208 ms	3 - 3 MB	NPU
DeepLabXception	QNN_DLC	float	Snapdragon® X Elite	22.689 ms	3 - 3 MB	NPU
DeepLabXception	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	15.977 ms	0 - 365 MB	NPU
DeepLabXception	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	113.17 ms	0 - 244 MB	NPU
DeepLabXception	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	21.99 ms	3 - 5 MB	NPU
DeepLabXception	QNN_DLC	float	Qualcomm® SA8775P	30.587 ms	1 - 245 MB	NPU
DeepLabXception	QNN_DLC	float	Qualcomm® QCS9075	35.269 ms	3 - 8 MB	NPU
DeepLabXception	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	45.66 ms	0 - 360 MB	NPU
DeepLabXception	QNN_DLC	float	Qualcomm® SA7255P	113.17 ms	0 - 244 MB	NPU
DeepLabXception	QNN_DLC	float	Qualcomm® SA8295P	38.892 ms	0 - 242 MB	NPU
DeepLabXception	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	12.197 ms	0 - 252 MB	NPU
DeepLabXception	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	9.228 ms	3 - 258 MB	NPU
DeepLabXception	QNN_DLC	w8a8	Snapdragon® X2 Elite	3.85 ms	1 - 1 MB	NPU
DeepLabXception	QNN_DLC	w8a8	Snapdragon® X Elite	8.543 ms	1 - 1 MB	NPU
DeepLabXception	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	5.561 ms	1 - 373 MB	NPU
DeepLabXception	QNN_DLC	w8a8	Qualcomm® QCS6490	33.733 ms	1 - 3 MB	NPU
DeepLabXception	QNN_DLC	w8a8	Qualcomm® QCS8275 (Proxy)	20.353 ms	1 - 282 MB	NPU
DeepLabXception	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	7.912 ms	1 - 212 MB	NPU
DeepLabXception	QNN_DLC	w8a8	Qualcomm® SA8775P	8.096 ms	1 - 284 MB	NPU
DeepLabXception	QNN_DLC	w8a8	Qualcomm® QCS9075	8.609 ms	1 - 3 MB	NPU
DeepLabXception	QNN_DLC	w8a8	Qualcomm® QCM6690	121.601 ms	1 - 303 MB	NPU
DeepLabXception	QNN_DLC	w8a8	Qualcomm® QCS8450 (Proxy)	13.818 ms	1 - 369 MB	NPU
DeepLabXception	QNN_DLC	w8a8	Qualcomm® SA7255P	20.353 ms	1 - 282 MB	NPU
DeepLabXception	QNN_DLC	w8a8	Qualcomm® SA8295P	12.655 ms	1 - 276 MB	NPU
DeepLabXception	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	4.362 ms	0 - 275 MB	NPU
DeepLabXception	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	12.1 ms	1 - 277 MB	NPU
DeepLabXception	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	3.264 ms	1 - 275 MB	NPU
DeepLabXception	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	16.186 ms	0 - 484 MB	NPU
DeepLabXception	TFLITE	float	Qualcomm® QCS8275 (Proxy)	113.246 ms	0 - 358 MB	NPU
DeepLabXception	TFLITE	float	Qualcomm® QCS8550 (Proxy)	21.206 ms	0 - 3 MB	NPU
DeepLabXception	TFLITE	float	Qualcomm® SA8775P	30.539 ms	0 - 358 MB	NPU
DeepLabXception	TFLITE	float	Qualcomm® QCS9075	35.04 ms	0 - 93 MB	NPU
DeepLabXception	TFLITE	float	Qualcomm® QCS8450 (Proxy)	46.675 ms	1 - 472 MB	NPU
DeepLabXception	TFLITE	float	Qualcomm® SA7255P	113.246 ms	0 - 358 MB	NPU
DeepLabXception	TFLITE	float	Qualcomm® SA8295P	38.653 ms	0 - 348 MB	NPU
DeepLabXception	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	12.199 ms	0 - 367 MB	NPU
DeepLabXception	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	9.22 ms	0 - 365 MB	NPU
DeepLabXception	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	4.958 ms	0 - 372 MB	NPU
DeepLabXception	TFLITE	w8a8	Qualcomm® QCS6490	32.146 ms	0 - 58 MB	NPU
DeepLabXception	TFLITE	w8a8	Qualcomm® QCS8275 (Proxy)	18.903 ms	0 - 284 MB	NPU
DeepLabXception	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	6.933 ms	0 - 3 MB	NPU
DeepLabXception	TFLITE	w8a8	Qualcomm® SA8775P	7.217 ms	0 - 285 MB	NPU
DeepLabXception	TFLITE	w8a8	Qualcomm® QCS9075	7.452 ms	0 - 47 MB	NPU
DeepLabXception	TFLITE	w8a8	Qualcomm® QCM6690	91.151 ms	0 - 295 MB	NPU
DeepLabXception	TFLITE	w8a8	Qualcomm® QCS8450 (Proxy)	11.984 ms	0 - 366 MB	NPU
DeepLabXception	TFLITE	w8a8	Qualcomm® SA7255P	18.903 ms	0 - 284 MB	NPU
DeepLabXception	TFLITE	w8a8	Qualcomm® SA8295P	11.552 ms	0 - 274 MB	NPU
DeepLabXception	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	3.918 ms	0 - 273 MB	NPU
DeepLabXception	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	11.81 ms	0 - 268 MB	NPU
DeepLabXception	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	2.899 ms	0 - 267 MB	NPU

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: -; Downloads are not tracked for this model. How to track

Paper for qualcomm/DeepLabXception

Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation

Paper • 1802.02611 • Published Feb 7, 2018