Hugging Face's logo

qualcomm
/
Depth-Anything

Depth Estimation
PyTorch
LiteRT
android
Model card Files
xet
Community
qaihm-bot commited on
Commit
c95d043
·
verified ·
1 Parent(s): 3a87ed2

v0.30.5

Browse files

See https://github.com/quic/ai-hub-models/releases/v0.30.5 for changelog.

Files changed (1) hide show
  1. README.md +33 -33
README.md CHANGED
@@ -34,35 +34,35 @@ More details on model performance across various devices, can be found
34
 
35
  | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
36
  |---|---|---|---|---|---|---|---|---|
37
- | Depth-Anything | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 274.831 ms | 1 - 585 MB | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
38
- | Depth-Anything | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 272.771 ms | 2 - 11 MB | NPU | Use Export Script |
39
- | Depth-Anything | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 180.679 ms | 1 - 621 MB | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
40
- | Depth-Anything | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 315.717 ms | 3 - 661 MB | NPU | Use Export Script |
41
- | Depth-Anything | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 137.612 ms | 1 - 92 MB | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
42
- | Depth-Anything | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 143.701 ms | 3 - 6 MB | NPU | Use Export Script |
43
- | Depth-Anything | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 142.018 ms | 1 - 585 MB | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
44
- | Depth-Anything | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 143.248 ms | 2 - 12 MB | NPU | Use Export Script |
45
- | Depth-Anything | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 274.831 ms | 1 - 585 MB | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
46
- | Depth-Anything | float | SA7255P ADP | Qualcomm® SA7255P | QNN | 272.771 ms | 2 - 11 MB | NPU | Use Export Script |
47
- | Depth-Anything | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 146.486 ms | 1 - 95 MB | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
48
- | Depth-Anything | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 140.508 ms | 3 - 5 MB | NPU | Use Export Script |
49
- | Depth-Anything | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 192.411 ms | 1 - 589 MB | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
50
- | Depth-Anything | float | SA8295P ADP | Qualcomm® SA8295P | QNN | 177.139 ms | 0 - 18 MB | NPU | Use Export Script |
51
- | Depth-Anything | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 146.914 ms | 0 - 98 MB | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
52
- | Depth-Anything | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 137.443 ms | 3 - 5 MB | NPU | Use Export Script |
53
- | Depth-Anything | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 142.018 ms | 1 - 585 MB | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
54
- | Depth-Anything | float | SA8775P ADP | Qualcomm® SA8775P | QNN | 143.248 ms | 2 - 12 MB | NPU | Use Export Script |
55
- | Depth-Anything | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 137.757 ms | 1 - 95 MB | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
56
- | Depth-Anything | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 145.698 ms | 0 - 108 MB | NPU | Use Export Script |
57
- | Depth-Anything | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 146.122 ms | 2 - 285 MB | NPU | [Depth-Anything.onnx](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.onnx) |
58
- | Depth-Anything | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 109.564 ms | 1 - 574 MB | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
59
- | Depth-Anything | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 109.812 ms | 3 - 652 MB | NPU | Use Export Script |
60
- | Depth-Anything | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 111.779 ms | 3 - 741 MB | NPU | [Depth-Anything.onnx](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.onnx) |
61
- | Depth-Anything | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 96.094 ms | 1 - 551 MB | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
62
- | Depth-Anything | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 89.508 ms | 0 - 670 MB | NPU | Use Export Script |
63
- | Depth-Anything | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 90.171 ms | 5 - 749 MB | NPU | [Depth-Anything.onnx](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.onnx) |
64
- | Depth-Anything | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 133.325 ms | 3 - 3 MB | NPU | Use Export Script |
65
- | Depth-Anything | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 147.396 ms | 66 - 66 MB | NPU | [Depth-Anything.onnx](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.onnx) |
66
 
67
 
68
 
@@ -126,8 +126,8 @@ Profiling Results
126
  Depth-Anything
127
  Device : cs_8275 (ANDROID 14)
128
  Runtime : TFLITE
129
- Estimated inference time (ms) : 274.8
130
- Estimated peak memory usage (MB): [1, 585]
131
  Total # Ops : 646
132
  Compute Unit(s) : npu (646 ops) gpu (0 ops) cpu (0 ops)
133
  ```
@@ -216,13 +216,13 @@ AI Hub. [Sign up for access](https://myaccount.qualcomm.com/signup).
216
  You can also run the demo on-device.
217
 
218
  ```bash
219
- python -m qai_hub_models.models.depth_anything.demo --on-device
220
  ```
221
 
222
  **NOTE**: If you want running in a Jupyter Notebook or Google Colab like
223
  environment, please add the following to your cell (instead of the above).
224
  ```
225
- %run -m qai_hub_models.models.depth_anything.demo -- --on-device
226
  ```
227
 
228
 
 
34
 
35
  | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
36
  |---|---|---|---|---|---|---|---|---|
37
+ | Depth-Anything | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 889.252 ms | 1 - 595 MB | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
38
+ | Depth-Anything | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 243.727 ms | 1 - 10 MB | NPU | Use Export Script |
39
+ | Depth-Anything | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 176.254 ms | 1 - 629 MB | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
40
+ | Depth-Anything | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 248.72 ms | 1 - 644 MB | NPU | Use Export Script |
41
+ | Depth-Anything | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 148.062 ms | 1 - 91 MB | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
42
+ | Depth-Anything | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 132.349 ms | 3 - 5 MB | NPU | Use Export Script |
43
+ | Depth-Anything | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 143.66 ms | 1 - 595 MB | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
44
+ | Depth-Anything | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 129.397 ms | 1 - 11 MB | NPU | Use Export Script |
45
+ | Depth-Anything | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 889.252 ms | 1 - 595 MB | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
46
+ | Depth-Anything | float | SA7255P ADP | Qualcomm® SA7255P | QNN | 243.727 ms | 1 - 10 MB | NPU | Use Export Script |
47
+ | Depth-Anything | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 147.541 ms | 1 - 100 MB | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
48
+ | Depth-Anything | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 132.594 ms | 3 - 6 MB | NPU | Use Export Script |
49
+ | Depth-Anything | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 182.658 ms | 1 - 598 MB | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
50
+ | Depth-Anything | float | SA8295P ADP | Qualcomm® SA8295P | QNN | 161.899 ms | 0 - 16 MB | NPU | Use Export Script |
51
+ | Depth-Anything | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 139.75 ms | 1 - 96 MB | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
52
+ | Depth-Anything | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 132.531 ms | 3 - 5 MB | NPU | Use Export Script |
53
+ | Depth-Anything | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 143.66 ms | 1 - 595 MB | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
54
+ | Depth-Anything | float | SA8775P ADP | Qualcomm® SA8775P | QNN | 129.397 ms | 1 - 11 MB | NPU | Use Export Script |
55
+ | Depth-Anything | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 150.727 ms | 1 - 96 MB | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
56
+ | Depth-Anything | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 130.352 ms | 3 - 97 MB | NPU | Use Export Script |
57
+ | Depth-Anything | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 135.102 ms | 4 - 206 MB | NPU | [Depth-Anything.onnx](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.onnx) |
58
+ | Depth-Anything | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 110.701 ms | 0 - 588 MB | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
59
+ | Depth-Anything | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 96.998 ms | 3 - 689 MB | NPU | Use Export Script |
60
+ | Depth-Anything | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 108.393 ms | 5 - 781 MB | NPU | [Depth-Anything.onnx](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.onnx) |
61
+ | Depth-Anything | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 98.053 ms | 1 - 562 MB | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
62
+ | Depth-Anything | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 85.768 ms | 2 - 699 MB | NPU | Use Export Script |
63
+ | Depth-Anything | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 88.674 ms | 5 - 769 MB | NPU | [Depth-Anything.onnx](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.onnx) |
64
+ | Depth-Anything | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 120.388 ms | 3 - 3 MB | NPU | Use Export Script |
65
+ | Depth-Anything | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 140.825 ms | 67 - 67 MB | NPU | [Depth-Anything.onnx](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.onnx) |
66
 
67
 
68
 
 
126
  Depth-Anything
127
  Device : cs_8275 (ANDROID 14)
128
  Runtime : TFLITE
129
+ Estimated inference time (ms) : 889.3
130
+ Estimated peak memory usage (MB): [1, 595]
131
  Total # Ops : 646
132
  Compute Unit(s) : npu (646 ops) gpu (0 ops) cpu (0 ops)
133
  ```
 
216
  You can also run the demo on-device.
217
 
218
  ```bash
219
+ python -m qai_hub_models.models.depth_anything.demo --eval-mode on-device
220
  ```
221
 
222
  **NOTE**: If you want running in a Jupyter Notebook or Google Colab like
223
  environment, please add the following to your cell (instead of the above).
224
  ```
225
+ %run -m qai_hub_models.models.depth_anything.demo -- --eval-mode on-device
226
  ```
227
 
228