Hugging Face's logo

qualcomm
/
DeepLabV3-Plus-MobileNet

Image Segmentation
PyTorch
android
Model card Files
xet
 Community
DeepLabV3-Plus-MobileNet
File size: 18,112 Bytes
83c145e
 
84b8214
83c145e
 
0b60cb3
83c145e
 
 
 
 
a897a3b
af6b6f7
83c145e
 
a897a3b
59d0495
a897a3b
 
 
 
 
 
 
 
 
 
 
 
59d0495
 
 
 
 
 
 
 
a897a3b
 
 
 
 
 
59d0495
a897a3b
 
 
 
 
 
59d0495
a897a3b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
59d0495
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1b36d5c
59d0495
1b36d5c
59d0495
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
dbabf92
83c145e
1af0e40
 
dbabf92
83c145e
 
 
 
 
3a34c61
83c145e
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
---
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).