Hugging Face's logo

qualcomm
/
EdgeTAM

Image Segmentation
PyTorch
foundation
android
Model card Files
xet
Community
qaihm-bot commited on
Commit
ae79e10
·
verified ·
1 Parent(s): 6885516

v0.47.0

Browse files

See https://github.com/quic/ai-hub-models/releases/v0.47.0 for changelog.

Files changed (2) hide show
  1. LICENSE +1 -0
  2. README.md +208 -0
LICENSE ADDED
@@ -0,0 +1 @@
 
 
1
+ The license of the original trained model can be found at https://github.com/facebookresearch/EdgeTAM/blob/main/LICENSE.
README.md ADDED
@@ -0,0 +1,208 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: pytorch
3
+ license: other
4
+ tags:
5
+ - foundation
6
+ - android
7
+ pipeline_tag: image-segmentation
8
+
9
+ ---
10
+
11
+ ![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/edgetam/web-assets/model_demo.png)
12
+
13
+ # EdgeTAM: Optimized for Qualcomm Devices
14
+
15
+ EdgeTAM is a lightweight and efficient version of SAM 2, optimized for on-device tracking and segmenting any object in videos and low memory usage by using a novel 2D Spatial Perceiver and RepViT backbone.
16
+
17
+ This is based on the implementation of EdgeTAM found [here](https://github.com/facebookresearch/EdgeTAM).
18
+ This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/edgetam) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
19
+
20
+ 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.
21
+
22
+ ## Getting Started
23
+ There are two ways to deploy this model on your device:
24
+
25
+ ### Option 1: Download Pre-Exported Models
26
+
27
+ Below are pre-exported model assets ready for deployment.
28
+
29
+ | Runtime | Precision | Chipset | SDK Versions | Download |
30
+ |---|---|---|---|---|
31
+ | ONNX | float | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/edgetam/releases/v0.47.0/edgetam-onnx-float.zip)
32
+ | ONNX | w8a8 | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/edgetam/releases/v0.47.0/edgetam-onnx-w8a8.zip)
33
+ | QNN_DLC | float | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/edgetam/releases/v0.47.0/edgetam-qnn_dlc-float.zip)
34
+ | QNN_DLC | w8a8 | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/edgetam/releases/v0.47.0/edgetam-qnn_dlc-w8a8.zip)
35
+ | TFLITE | float | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/edgetam/releases/v0.47.0/edgetam-tflite-float.zip)
36
+ | TFLITE | w8a8 | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/edgetam/releases/v0.47.0/edgetam-tflite-w8a8.zip)
37
+
38
+ For more device-specific assets and performance metrics, visit **[EdgeTAM on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/edgetam)**.
39
+
40
+
41
+ ### Option 2: Export with Custom Configurations
42
+
43
+ Use the [Qualcomm® AI Hub Models](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/edgetam) Python library to compile and export the model with your own:
44
+ - Custom weights (e.g., fine-tuned checkpoints)
45
+ - Custom input shapes
46
+ - Target device and runtime configurations
47
+
48
+ This option is ideal if you need to customize the model beyond the default configuration provided here.
49
+
50
+ See our repository for [EdgeTAM on GitHub](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/edgetam) for usage instructions.
51
+
52
+ ## Model Details
53
+
54
+ **Model Type:** Model_use_case.semantic_segmentation
55
+
56
+ **Model Stats:**
57
+ - Model checkpoint: edgetam.pt
58
+ - Input resolution (Encoder): 1024x1024
59
+ - Number of parameters (EdgeTAMEncoder): 8.3M
60
+ - Model size (EdgeTAMEncoder) (float): 33 MB
61
+ - Number of parameters (EdgeTAMDecoder): 6.22M
62
+ - Model size (EdgeTAMDecoder) (float): 23.7 MB
63
+
64
+ ## Performance Summary
65
+ | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
66
+ |---|---|---|---|---|---|---
67
+ | EdgeTAMDecoder | ONNX | float | Snapdragon® X Elite | 6.843 ms | 16 - 16 MB | NPU
68
+ | EdgeTAMDecoder | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 4.706 ms | 17 - 334 MB | NPU
69
+ | EdgeTAMDecoder | ONNX | float | Qualcomm® QCS8550 (Proxy) | 6.785 ms | 16 - 24 MB | NPU
70
+ | EdgeTAMDecoder | ONNX | float | Qualcomm® QCS9075 | 8.422 ms | 16 - 19 MB | NPU
71
+ | EdgeTAMDecoder | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 3.491 ms | 2 - 236 MB | NPU
72
+ | EdgeTAMDecoder | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 3.15 ms | 0 - 228 MB | NPU
73
+ | EdgeTAMDecoder | ONNX | float | Snapdragon® X2 Elite | 3.303 ms | 15 - 15 MB | NPU
74
+ | EdgeTAMDecoder | ONNX | w8a8 | Snapdragon® X Elite | 12.58 ms | 4 - 4 MB | NPU
75
+ | EdgeTAMDecoder | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 9.225 ms | 4 - 275 MB | NPU
76
+ | EdgeTAMDecoder | ONNX | w8a8 | Qualcomm® QCS6490 | 127.159 ms | 47 - 58 MB | CPU
77
+ | EdgeTAMDecoder | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 12.466 ms | 2 - 10 MB | NPU
78
+ | EdgeTAMDecoder | ONNX | w8a8 | Qualcomm® QCS9075 | 14.219 ms | 4 - 7 MB | NPU
79
+ | EdgeTAMDecoder | ONNX | w8a8 | Qualcomm® QCM6690 | 58.694 ms | 44 - 66 MB | CPU
80
+ | EdgeTAMDecoder | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 6.716 ms | 2 - 243 MB | NPU
81
+ | EdgeTAMDecoder | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 51.163 ms | 50 - 68 MB | CPU
82
+ | EdgeTAMDecoder | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 5.307 ms | 4 - 271 MB | NPU
83
+ | EdgeTAMDecoder | ONNX | w8a8 | Snapdragon® X2 Elite | 5.449 ms | 4 - 4 MB | NPU
84
+ | EdgeTAMDecoder | QNN_DLC | float | Snapdragon® X Elite | 5.462 ms | 16 - 16 MB | NPU
85
+ | EdgeTAMDecoder | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 3.6 ms | 3 - 234 MB | NPU
86
+ | EdgeTAMDecoder | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 12.442 ms | 16 - 281 MB | NPU
87
+ | EdgeTAMDecoder | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 5.255 ms | 16 - 45 MB | NPU
88
+ | EdgeTAMDecoder | QNN_DLC | float | Qualcomm® SA8775P | 6.494 ms | 16 - 217 MB | NPU
89
+ | EdgeTAMDecoder | QNN_DLC | float | Qualcomm® QCS9075 | 6.873 ms | 18 - 36 MB | NPU
90
+ | EdgeTAMDecoder | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 8.859 ms | 4 - 226 MB | NPU
91
+ | EdgeTAMDecoder | QNN_DLC | float | Qualcomm® SA7255P | 12.442 ms | 16 - 281 MB | NPU
92
+ | EdgeTAMDecoder | QNN_DLC | float | Qualcomm® SA8295P | 8.061 ms | 0 - 215 MB | NPU
93
+ | EdgeTAMDecoder | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 2.904 ms | 4 - 213 MB | NPU
94
+ | EdgeTAMDecoder | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 2.53 ms | 16 - 214 MB | NPU
95
+ | EdgeTAMDecoder | QNN_DLC | float | Snapdragon® X2 Elite | 2.999 ms | 16 - 16 MB | NPU
96
+ | EdgeTAMDecoder | QNN_DLC | w8a8 | Snapdragon® X Elite | 2.253 ms | 4 - 4 MB | NPU
97
+ | EdgeTAMDecoder | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 1.298 ms | 0 - 102 MB | NPU
98
+ | EdgeTAMDecoder | QNN_DLC | w8a8 | Qualcomm® QCS8275 (Proxy) | 3.866 ms | 4 - 83 MB | NPU
99
+ | EdgeTAMDecoder | QNN_DLC | w8a8 | Qualcomm® QCS8550 (Proxy) | 1.912 ms | 0 - 31 MB | NPU
100
+ | EdgeTAMDecoder | QNN_DLC | w8a8 | Qualcomm® SA8775P | 9.278 ms | 4 - 83 MB | NPU
101
+ | EdgeTAMDecoder | QNN_DLC | w8a8 | Qualcomm® QCS9075 | 2.475 ms | 1 - 7 MB | NPU
102
+ | EdgeTAMDecoder | QNN_DLC | w8a8 | Qualcomm® QCM6690 | 9.968 ms | 4 - 264 MB | NPU
103
+ | EdgeTAMDecoder | QNN_DLC | w8a8 | Qualcomm® QCS8450 (Proxy) | 2.65 ms | 0 - 102 MB | NPU
104
+ | EdgeTAMDecoder | QNN_DLC | w8a8 | Qualcomm® SA7255P | 3.866 ms | 4 - 83 MB | NPU
105
+ | EdgeTAMDecoder | QNN_DLC | w8a8 | Qualcomm® SA8295P | 2.964 ms | 4 - 97 MB | NPU
106
+ | EdgeTAMDecoder | QNN_DLC | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 1.035 ms | 1 - 83 MB | NPU
107
+ | EdgeTAMDecoder | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 2.62 ms | 4 - 208 MB | NPU
108
+ | EdgeTAMDecoder | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.894 ms | 1 - 106 MB | NPU
109
+ | EdgeTAMDecoder | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 1.367 ms | 4 - 4 MB | NPU
110
+ | EdgeTAMDecoder | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 3.705 ms | 0 - 244 MB | NPU
111
+ | EdgeTAMDecoder | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 12.818 ms | 0 - 210 MB | NPU
112
+ | EdgeTAMDecoder | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 5.359 ms | 0 - 7 MB | NPU
113
+ | EdgeTAMDecoder | TFLITE | float | Qualcomm® SA8775P | 6.772 ms | 0 - 211 MB | NPU
114
+ | EdgeTAMDecoder | TFLITE | float | Qualcomm® QCS9075 | 6.998 ms | 0 - 30 MB | NPU
115
+ | EdgeTAMDecoder | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 8.896 ms | 0 - 229 MB | NPU
116
+ | EdgeTAMDecoder | TFLITE | float | Qualcomm® SA7255P | 12.818 ms | 0 - 210 MB | NPU
117
+ | EdgeTAMDecoder | TFLITE | float | Qualcomm® SA8295P | 8.043 ms | 0 - 222 MB | NPU
118
+ | EdgeTAMDecoder | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 2.899 ms | 0 - 218 MB | NPU
119
+ | EdgeTAMDecoder | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 2.542 ms | 0 - 206 MB | NPU
120
+ | EdgeTAMDecoder | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 9.195 ms | 0 - 261 MB | NPU
121
+ | EdgeTAMDecoder | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 20.937 ms | 0 - 212 MB | NPU
122
+ | EdgeTAMDecoder | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 12.191 ms | 0 - 2 MB | NPU
123
+ | EdgeTAMDecoder | TFLITE | w8a8 | Qualcomm® SA8775P | 51.678 ms | 0 - 213 MB | NPU
124
+ | EdgeTAMDecoder | TFLITE | w8a8 | Qualcomm® QCS9075 | 13.187 ms | 0 - 14 MB | NPU
125
+ | EdgeTAMDecoder | TFLITE | w8a8 | Qualcomm® QCM6690 | 51.184 ms | 12 - 205 MB | NPU
126
+ | EdgeTAMDecoder | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 15.573 ms | 0 - 239 MB | NPU
127
+ | EdgeTAMDecoder | TFLITE | w8a8 | Qualcomm® SA7255P | 20.937 ms | 0 - 212 MB | NPU
128
+ | EdgeTAMDecoder | TFLITE | w8a8 | Qualcomm® SA8295P | 14.243 ms | 0 - 223 MB | NPU
129
+ | EdgeTAMDecoder | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 6.872 ms | 0 - 225 MB | NPU
130
+ | EdgeTAMDecoder | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 16.171 ms | 12 - 208 MB | NPU
131
+ | EdgeTAMDecoder | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 5.415 ms | 0 - 239 MB | NPU
132
+ | EdgeTAMEncoder | ONNX | float | Snapdragon® X Elite | 20.284 ms | 34 - 34 MB | NPU
133
+ | EdgeTAMEncoder | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 14.878 ms | 45 - 396 MB | NPU
134
+ | EdgeTAMEncoder | ONNX | float | Qualcomm® QCS8550 (Proxy) | 20.231 ms | 30 - 47 MB | NPU
135
+ | EdgeTAMEncoder | ONNX | float | Qualcomm® QCS9075 | 28.941 ms | 12 - 27 MB | NPU
136
+ | EdgeTAMEncoder | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 9.871 ms | 0 - 264 MB | NPU
137
+ | EdgeTAMEncoder | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 7.757 ms | 45 - 328 MB | NPU
138
+ | EdgeTAMEncoder | ONNX | float | Snapdragon® X2 Elite | 8.844 ms | 32 - 32 MB | NPU
139
+ | EdgeTAMEncoder | ONNX | w8a8 | Snapdragon® X Elite | 8.85 ms | 16 - 16 MB | NPU
140
+ | EdgeTAMEncoder | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 5.462 ms | 13 - 321 MB | NPU
141
+ | EdgeTAMEncoder | ONNX | w8a8 | Qualcomm® QCS6490 | 488.296 ms | 203 - 213 MB | CPU
142
+ | EdgeTAMEncoder | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 8.221 ms | 11 - 17 MB | NPU
143
+ | EdgeTAMEncoder | ONNX | w8a8 | Qualcomm® QCS9075 | 9.524 ms | 12 - 15 MB | NPU
144
+ | EdgeTAMEncoder | ONNX | w8a8 | Qualcomm® QCM6690 | 323.926 ms | 141 - 152 MB | CPU
145
+ | EdgeTAMEncoder | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 4.476 ms | 10 - 273 MB | NPU
146
+ | EdgeTAMEncoder | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 308.903 ms | 207 - 219 MB | CPU
147
+ | EdgeTAMEncoder | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 3.303 ms | 0 - 275 MB | NPU
148
+ | EdgeTAMEncoder | ONNX | w8a8 | Snapdragon® X2 Elite | 3.491 ms | 18 - 18 MB | NPU
149
+ | EdgeTAMEncoder | QNN_DLC | float | Snapdragon® X Elite | 24.17 ms | 12 - 12 MB | NPU
150
+ | EdgeTAMEncoder | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 16.948 ms | 12 - 368 MB | NPU
151
+ | EdgeTAMEncoder | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 69.756 ms | 1 - 261 MB | NPU
152
+ | EdgeTAMEncoder | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 23.673 ms | 12 - 14 MB | NPU
153
+ | EdgeTAMEncoder | QNN_DLC | float | Qualcomm® SA8775P | 28.167 ms | 0 - 261 MB | NPU
154
+ | EdgeTAMEncoder | QNN_DLC | float | Qualcomm® QCS9075 | 34.311 ms | 12 - 42 MB | NPU
155
+ | EdgeTAMEncoder | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 98.559 ms | 12 - 466 MB | NPU
156
+ | EdgeTAMEncoder | QNN_DLC | float | Qualcomm® SA7255P | 69.756 ms | 1 - 261 MB | NPU
157
+ | EdgeTAMEncoder | QNN_DLC | float | Qualcomm® SA8295P | 57.556 ms | 0 - 358 MB | NPU
158
+ | EdgeTAMEncoder | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 11.891 ms | 12 - 290 MB | NPU
159
+ | EdgeTAMEncoder | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 8.637 ms | 12 - 308 MB | NPU
160
+ | EdgeTAMEncoder | QNN_DLC | float | Snapdragon® X2 Elite | 10.279 ms | 12 - 12 MB | NPU
161
+ | EdgeTAMEncoder | QNN_DLC | w8a8 | Snapdragon® X Elite | 12.205 ms | 3 - 3 MB | NPU
162
+ | EdgeTAMEncoder | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 7.566 ms | 3 - 317 MB | NPU
163
+ | EdgeTAMEncoder | QNN_DLC | w8a8 | Qualcomm® QCS8275 (Proxy) | 20.524 ms | 3 - 266 MB | NPU
164
+ | EdgeTAMEncoder | QNN_DLC | w8a8 | Qualcomm® QCS8550 (Proxy) | 11.404 ms | 3 - 160 MB | NPU
165
+ | EdgeTAMEncoder | QNN_DLC | w8a8 | Qualcomm® SA8775P | 12.088 ms | 3 - 266 MB | NPU
166
+ | EdgeTAMEncoder | QNN_DLC | w8a8 | Qualcomm® QCS9075 | 15.141 ms | 3 - 12 MB | NPU
167
+ | EdgeTAMEncoder | QNN_DLC | w8a8 | Qualcomm® QCM6690 | 59.198 ms | 3 - 269 MB | NPU
168
+ | EdgeTAMEncoder | QNN_DLC | w8a8 | Qualcomm® QCS8450 (Proxy) | 1963.093 ms | 3 - 209 MB | NPU
169
+ | EdgeTAMEncoder | QNN_DLC | w8a8 | Qualcomm® SA7255P | 20.524 ms | 3 - 266 MB | NPU
170
+ | EdgeTAMEncoder | QNN_DLC | w8a8 | Qualcomm® SA8295P | 2138.343 ms | 3 - 161 MB | NPU
171
+ | EdgeTAMEncoder | QNN_DLC | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 5.362 ms | 0 - 255 MB | NPU
172
+ | EdgeTAMEncoder | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 12.383 ms | 3 - 301 MB | NPU
173
+ | EdgeTAMEncoder | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 4.229 ms | 3 - 269 MB | NPU
174
+ | EdgeTAMEncoder | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 4.859 ms | 3 - 3 MB | NPU
175
+ | EdgeTAMEncoder | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 16.568 ms | 15 - 372 MB | NPU
176
+ | EdgeTAMEncoder | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 68.178 ms | 16 - 283 MB | NPU
177
+ | EdgeTAMEncoder | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 22.855 ms | 16 - 18 MB | NPU
178
+ | EdgeTAMEncoder | TFLITE | float | Qualcomm® SA8775P | 27.351 ms | 16 - 280 MB | NPU
179
+ | EdgeTAMEncoder | TFLITE | float | Qualcomm® QCS9075 | 32.802 ms | 0 - 42 MB | NPU
180
+ | EdgeTAMEncoder | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 99.118 ms | 16 - 460 MB | NPU
181
+ | EdgeTAMEncoder | TFLITE | float | Qualcomm® SA7255P | 68.178 ms | 16 - 283 MB | NPU
182
+ | EdgeTAMEncoder | TFLITE | float | Qualcomm® SA8295P | 54.228 ms | 16 - 366 MB | NPU
183
+ | EdgeTAMEncoder | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 11.455 ms | 14 - 295 MB | NPU
184
+ | EdgeTAMEncoder | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 8.409 ms | 16 - 318 MB | NPU
185
+ | EdgeTAMEncoder | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 4.806 ms | 4 - 313 MB | NPU
186
+ | EdgeTAMEncoder | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 13.791 ms | 4 - 258 MB | NPU
187
+ | EdgeTAMEncoder | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 7.021 ms | 4 - 198 MB | NPU
188
+ | EdgeTAMEncoder | TFLITE | w8a8 | Qualcomm® SA8775P | 7.862 ms | 4 - 259 MB | NPU
189
+ | EdgeTAMEncoder | TFLITE | w8a8 | Qualcomm® QCS9075 | 7.867 ms | 4 - 19 MB | NPU
190
+ | EdgeTAMEncoder | TFLITE | w8a8 | Qualcomm® QCM6690 | 70.404 ms | 4 - 237 MB | NPU
191
+ | EdgeTAMEncoder | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 2256.068 ms | 0 - 330 MB | NPU
192
+ | EdgeTAMEncoder | TFLITE | w8a8 | Qualcomm® SA7255P | 13.791 ms | 4 - 258 MB | NPU
193
+ | EdgeTAMEncoder | TFLITE | w8a8 | Qualcomm® SA8295P | 2221.179 ms | 0 - 277 MB | NPU
194
+ | EdgeTAMEncoder | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 2.996 ms | 2 - 250 MB | NPU
195
+ | EdgeTAMEncoder | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 8.263 ms | 3 - 262 MB | NPU
196
+ | EdgeTAMEncoder | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 2.662 ms | 4 - 262 MB | NPU
197
+
198
+ ## License
199
+ * The license for the original implementation of EdgeTAM can be found
200
+ [here](https://github.com/facebookresearch/EdgeTAM/blob/main/LICENSE).
201
+
202
+ ## References
203
+ * [EdgeTAM: On-Device Track Anything Model](https://arxiv.org/abs/2501.07256)
204
+ * [Source Model Implementation](https://github.com/facebookresearch/EdgeTAM)
205
+
206
+ ## Community
207
+ * Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI.
208
+ * For questions or feedback please [reach out to us](mailto:ai-hub-support@qti.qualcomm.com).