Yuantao Feng
commited on
Commit
·
ed7f07e
1
Parent(s):
23d8387
bring the table of benchmark results to the front (#17)
Browse files
README.md
CHANGED
|
@@ -14,18 +14,6 @@ Guidelines:
|
|
| 14 |
|
| 15 |
## Models & Benchmarks
|
| 16 |
|
| 17 |
-
Hardware Setup:
|
| 18 |
-
- `CPU x86_64`: INTEL CPU i7-5930K @ 3.50GHz, 6 cores, 12 threads.
|
| 19 |
-
- `CPU ARM`: Raspberry 4B, BCM2711B0 @ 1.5GHz (Cortex A-72), 4 cores, 4 threads.
|
| 20 |
-
- `GPU CUDA`: NVIDIA Jetson Nano B01, 128-core Maxwell, Quad-core ARM A57 @ 1.43 GHz.
|
| 21 |
-
|
| 22 |
-
***Important Notes***:
|
| 23 |
-
- The time data that shown on the following table presents the time elapsed from preprocess (resize is excluded), to a forward pass of a network, and postprocess to get final results.
|
| 24 |
-
- The time data that shown on the following table is the median of 10 runs. Different metrics may be applied to some specific models.
|
| 25 |
-
- Batch size is 1 for all benchmark results.
|
| 26 |
-
- View [benchmark/config](./benchmark/config) for more details on benchmarking different models.
|
| 27 |
-
- `---` means this model is not availble to run on the device.
|
| 28 |
-
|
| 29 |
| Model | Input Size | CPU x86_64 (ms) | CPU ARM (ms) | GPU CUDA (ms) |
|
| 30 |
|-------|------------|-----------------|--------------|---------------|
|
| 31 |
| [YuNet](./models/face_detection_yunet) | 160x120 | 1.45 | 6.22 | 12.18 |
|
|
@@ -38,6 +26,18 @@ Hardware Setup:
|
|
| 38 |
| [WeChatQRCode](./models/qrcode_wechatqrcode) | 100x100 | 7.04 | 37.68 | --- |
|
| 39 |
| [DaSiamRPN](./models/object_tracking_dasiamrpn) | 1280x720 | 36.15 | 705.48 | 76.82 |
|
| 40 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 41 |
## License
|
| 42 |
|
| 43 |
OpenCV Zoo is licensed under the [Apache 2.0 license](./LICENSE). Please refer to licenses of different models.
|
|
|
|
| 14 |
|
| 15 |
## Models & Benchmarks
|
| 16 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
| Model | Input Size | CPU x86_64 (ms) | CPU ARM (ms) | GPU CUDA (ms) |
|
| 18 |
|-------|------------|-----------------|--------------|---------------|
|
| 19 |
| [YuNet](./models/face_detection_yunet) | 160x120 | 1.45 | 6.22 | 12.18 |
|
|
|
|
| 26 |
| [WeChatQRCode](./models/qrcode_wechatqrcode) | 100x100 | 7.04 | 37.68 | --- |
|
| 27 |
| [DaSiamRPN](./models/object_tracking_dasiamrpn) | 1280x720 | 36.15 | 705.48 | 76.82 |
|
| 28 |
|
| 29 |
+
Hardware Setup:
|
| 30 |
+
- `CPU x86_64`: INTEL CPU i7-5930K @ 3.50GHz, 6 cores, 12 threads.
|
| 31 |
+
- `CPU ARM`: Raspberry 4B, BCM2711B0 @ 1.5GHz (Cortex A-72), 4 cores, 4 threads.
|
| 32 |
+
- `GPU CUDA`: NVIDIA Jetson Nano B01, 128-core Maxwell, Quad-core ARM A57 @ 1.43 GHz.
|
| 33 |
+
|
| 34 |
+
***Important Notes***:
|
| 35 |
+
- The time data that shown on the following table presents the time elapsed from preprocess (resize is excluded), to a forward pass of a network, and postprocess to get final results.
|
| 36 |
+
- The time data that shown on the following table is the median of 10 runs. Different metrics may be applied to some specific models.
|
| 37 |
+
- Batch size is 1 for all benchmark results.
|
| 38 |
+
- View [benchmark/config](./benchmark/config) for more details on benchmarking different models.
|
| 39 |
+
- `---` means this model is not availble to run on the device.
|
| 40 |
+
|
| 41 |
## License
|
| 42 |
|
| 43 |
OpenCV Zoo is licensed under the [Apache 2.0 license](./LICENSE). Please refer to licenses of different models.
|