Hugging Face's logo

qualcomm
/
Midas-V2

Depth Estimation
PyTorch
TF Lite
ONNX
android
Model card Files Community
qaihm-bot commited on
Commit
fb6c1d5
·
verified ·
1 Parent(s): dc6bc0c

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +30 -30
README.md CHANGED
@@ -34,45 +34,44 @@ More details on model performance across various devices, can be found
34
 
35
  | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
36
  |---|---|---|---|---|---|---|---|---|
37
- | Midas-V2 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 3.14 ms | 0 - 152 MB | FP16 | NPU | [Midas-V2.tflite](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.tflite) |
38
- | Midas-V2 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 3.281 ms | 0 - 101 MB | FP16 | NPU | [Midas-V2.so](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.so) |
39
- | Midas-V2 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 3.369 ms | 0 - 148 MB | FP16 | NPU | [Midas-V2.onnx](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.onnx) |
40
- | Midas-V2 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 2.209 ms | 0 - 28 MB | FP16 | NPU | [Midas-V2.tflite](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.tflite) |
41
- | Midas-V2 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 2.299 ms | 1 - 32 MB | FP16 | NPU | [Midas-V2.so](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.so) |
42
- | Midas-V2 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 2.375 ms | 0 - 37 MB | FP16 | NPU | [Midas-V2.onnx](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.onnx) |
43
- | Midas-V2 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 2.025 ms | 0 - 28 MB | FP16 | NPU | [Midas-V2.tflite](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.tflite) |
44
- | Midas-V2 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 2.114 ms | 0 - 27 MB | FP16 | NPU | Use Export Script |
45
- | Midas-V2 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 2.207 ms | 0 - 31 MB | FP16 | NPU | [Midas-V2.onnx](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.onnx) |
46
- | Midas-V2 | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 3.135 ms | 0 - 141 MB | FP16 | NPU | [Midas-V2.tflite](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.tflite) |
47
- | Midas-V2 | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 3.032 ms | 1 - 4 MB | FP16 | NPU | Use Export Script |
48
- | Midas-V2 | SA7255P ADP | SA7255P | TFLITE | 84.435 ms | 0 - 23 MB | FP16 | NPU | [Midas-V2.tflite](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.tflite) |
49
- | Midas-V2 | SA7255P ADP | SA7255P | QNN | 84.277 ms | 1 - 10 MB | FP16 | NPU | Use Export Script |
50
- | Midas-V2 | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 3.146 ms | 0 - 82 MB | FP16 | NPU | [Midas-V2.tflite](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.tflite) |
51
- | Midas-V2 | SA8255 (Proxy) | SA8255P Proxy | QNN | 3.04 ms | 1 - 4 MB | FP16 | NPU | Use Export Script |
52
- | Midas-V2 | SA8295P ADP | SA8295P | TFLITE | 5.614 ms | 0 - 25 MB | FP16 | NPU | [Midas-V2.tflite](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.tflite) |
53
- | Midas-V2 | SA8295P ADP | SA8295P | QNN | 5.552 ms | 1 - 15 MB | FP16 | NPU | Use Export Script |
54
- | Midas-V2 | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 3.132 ms | 0 - 161 MB | FP16 | NPU | [Midas-V2.tflite](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.tflite) |
55
- | Midas-V2 | SA8650 (Proxy) | SA8650P Proxy | QNN | 3.044 ms | 1 - 4 MB | FP16 | NPU | Use Export Script |
56
- | Midas-V2 | SA8775P ADP | SA8775P | TFLITE | 5.363 ms | 0 - 23 MB | FP16 | NPU | [Midas-V2.tflite](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.tflite) |
57
- | Midas-V2 | SA8775P ADP | SA8775P | QNN | 5.223 ms | 1 - 11 MB | FP16 | NPU | Use Export Script |
58
- | Midas-V2 | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 4.779 ms | 0 - 28 MB | FP16 | NPU | [Midas-V2.tflite](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.tflite) |
59
- | Midas-V2 | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 4.997 ms | 1 - 32 MB | FP16 | NPU | Use Export Script |
60
- | Midas-V2 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 3.184 ms | 1 - 1 MB | FP16 | NPU | Use Export Script |
61
- | Midas-V2 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 3.375 ms | 37 - 37 MB | FP16 | NPU | [Midas-V2.onnx](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.onnx) |
62
 
63
 
64
 
65
 
66
  ## Installation
67
 
68
- This model can be installed as a Python package via pip.
69
 
 
70
  ```bash
71
  pip install "qai-hub-models[midas]"
72
  ```
73
 
74
 
75
-
76
  ## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
77
 
78
  Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
@@ -124,7 +123,7 @@ Midas-V2
124
  Device : Samsung Galaxy S23 (13)
125
  Runtime : TFLITE
126
  Estimated inference time (ms) : 3.1
127
- Estimated peak memory usage (MB): [0, 152]
128
  Total # Ops : 138
129
  Compute Unit(s) : NPU (138 ops)
130
  ```
@@ -151,7 +150,7 @@ from qai_hub_models.models.midas import Model
151
  torch_model = Model.from_pretrained()
152
 
153
  # Device
154
- device = hub.Device("Samsung Galaxy S23")
155
 
156
  # Trace model
157
  input_shape = torch_model.get_input_spec()
@@ -243,7 +242,8 @@ Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
243
 
244
 
245
  ## License
246
- * The license for the original implementation of Midas-V2 can be found [here](https://github.com/isl-org/MiDaS/blob/master/LICENSE).
 
247
  * The license for the compiled assets for on-device deployment can be found [here](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/Qualcomm+AI+Hub+Proprietary+License.pdf)
248
 
249
 
 
34
 
35
  | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
36
  |---|---|---|---|---|---|---|---|---|
37
+ | Midas-V2 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 3.133 ms | 0 - 162 MB | FP16 | NPU | [Midas-V2.tflite](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.tflite) |
38
+ | Midas-V2 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 3.271 ms | 0 - 111 MB | FP16 | NPU | [Midas-V2.so](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.so) |
39
+ | Midas-V2 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 3.327 ms | 0 - 138 MB | FP16 | NPU | [Midas-V2.onnx](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.onnx) |
40
+ | Midas-V2 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 2.208 ms | 0 - 31 MB | FP16 | NPU | [Midas-V2.tflite](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.tflite) |
41
+ | Midas-V2 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 2.315 ms | 1 - 29 MB | FP16 | NPU | [Midas-V2.so](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.so) |
42
+ | Midas-V2 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 2.386 ms | 0 - 37 MB | FP16 | NPU | [Midas-V2.onnx](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.onnx) |
43
+ | Midas-V2 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 2.024 ms | 0 - 27 MB | FP16 | NPU | [Midas-V2.tflite](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.tflite) |
44
+ | Midas-V2 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 1.792 ms | 1 - 28 MB | FP16 | NPU | Use Export Script |
45
+ | Midas-V2 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 1.912 ms | 1 - 32 MB | FP16 | NPU | [Midas-V2.onnx](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.onnx) |
46
+ | Midas-V2 | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 3.141 ms | 0 - 141 MB | FP16 | NPU | [Midas-V2.tflite](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.tflite) |
47
+ | Midas-V2 | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 3.033 ms | 1 - 4 MB | FP16 | NPU | Use Export Script |
48
+ | Midas-V2 | SA7255P ADP | SA7255P | TFLITE | 84.338 ms | 0 - 23 MB | FP16 | NPU | [Midas-V2.tflite](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.tflite) |
49
+ | Midas-V2 | SA7255P ADP | SA7255P | QNN | 84.188 ms | 1 - 10 MB | FP16 | NPU | Use Export Script |
50
+ | Midas-V2 | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 3.148 ms | 0 - 151 MB | FP16 | NPU | [Midas-V2.tflite](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.tflite) |
51
+ | Midas-V2 | SA8255 (Proxy) | SA8255P Proxy | QNN | 3.048 ms | 1 - 4 MB | FP16 | NPU | Use Export Script |
52
+ | Midas-V2 | SA8295P ADP | SA8295P | TFLITE | 5.662 ms | 0 - 25 MB | FP16 | NPU | [Midas-V2.tflite](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.tflite) |
53
+ | Midas-V2 | SA8295P ADP | SA8295P | QNN | 5.628 ms | 1 - 15 MB | FP16 | NPU | Use Export Script |
54
+ | Midas-V2 | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 3.127 ms | 0 - 141 MB | FP16 | NPU | [Midas-V2.tflite](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.tflite) |
55
+ | Midas-V2 | SA8650 (Proxy) | SA8650P Proxy | QNN | 3.043 ms | 1 - 4 MB | FP16 | NPU | Use Export Script |
56
+ | Midas-V2 | SA8775P ADP | SA8775P | TFLITE | 5.312 ms | 0 - 23 MB | FP16 | NPU | [Midas-V2.tflite](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.tflite) |
57
+ | Midas-V2 | SA8775P ADP | SA8775P | QNN | 5.221 ms | 1 - 11 MB | FP16 | NPU | Use Export Script |
58
+ | Midas-V2 | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 4.76 ms | 0 - 29 MB | FP16 | NPU | [Midas-V2.tflite](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.tflite) |
59
+ | Midas-V2 | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 4.902 ms | 0 - 29 MB | FP16 | NPU | Use Export Script |
60
+ | Midas-V2 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 3.187 ms | 1 - 1 MB | FP16 | NPU | Use Export Script |
61
+ | Midas-V2 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 3.359 ms | 37 - 37 MB | FP16 | NPU | [Midas-V2.onnx](https://huggingface.co/qualcomm/Midas-V2/blob/main/Midas-V2.onnx) |
62
 
63
 
64
 
65
 
66
  ## Installation
67
 
 
68
 
69
+ Install the package via pip:
70
  ```bash
71
  pip install "qai-hub-models[midas]"
72
  ```
73
 
74
 
 
75
  ## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
76
 
77
  Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
 
123
  Device : Samsung Galaxy S23 (13)
124
  Runtime : TFLITE
125
  Estimated inference time (ms) : 3.1
126
+ Estimated peak memory usage (MB): [0, 162]
127
  Total # Ops : 138
128
  Compute Unit(s) : NPU (138 ops)
129
  ```
 
150
  torch_model = Model.from_pretrained()
151
 
152
  # Device
153
+ device = hub.Device("Samsung Galaxy S24")
154
 
155
  # Trace model
156
  input_shape = torch_model.get_input_spec()
 
242
 
243
 
244
  ## License
245
+ * The license for the original implementation of Midas-V2 can be found
246
+ [here](https://github.com/isl-org/MiDaS/blob/master/LICENSE).
247
  * The license for the compiled assets for on-device deployment can be found [here](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/Qualcomm+AI+Hub+Proprietary+License.pdf)
248
 
249