Hugging Face's logo

qualcomm
/
RTMDet

Object Detection
PyTorch
real_time
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
1a3e8be
·
verified ·
1 Parent(s): 6a29a65

v0.49.1

Browse files

See https://github.com/qualcomm/ai-hub-models/releases/v0.49.1 for changelog.

Files changed (1) hide show
  1. README.md +27 -27
README.md CHANGED
@@ -15,18 +15,18 @@ pipeline_tag: object-detection
15
  RTMDet is a highly efficient model for real-time object detection,capable of predicting both the bounding boxes and classes of objects within an image.It is highly optimized for real-time applications, making it reliable for industrial and commercial use
16
 
17
  This is based on the implementation of RTMDet found [here](https://github.com/open-mmlab/mmdetection/tree/3.x/configs/rtmdet).
18
- 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/main/qai_hub_models/models/rtmdet) 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
  Due to licensing restrictions, we cannot distribute pre-exported model assets for this model.
24
- Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/qai_hub_models/models/rtmdet) Python library to compile and export the model with your own:
25
  - Custom weights (e.g., fine-tuned checkpoints)
26
  - Custom input shapes
27
  - Target device and runtime configurations
28
 
29
- See our repository for [RTMDet on GitHub](https://github.com/qualcomm/ai-hub-models/blob/main/qai_hub_models/models/rtmdet) for usage instructions.
30
 
31
 
32
  ## Model Details
@@ -42,30 +42,30 @@ See our repository for [RTMDet on GitHub](https://github.com/qualcomm/ai-hub-mod
42
  ## Performance Summary
43
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
44
  |---|---|---|---|---|---|---
45
- | RTMDet | ONNX | float | Snapdragon® X2 Elite | 8.171 ms | 53 - 53 MB | NPU
46
- | RTMDet | ONNX | float | Snapdragon® X Elite | 14.218 ms | 51 - 51 MB | NPU
47
- | RTMDet | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 10.625 ms | 5 - 235 MB | NPU
48
- | RTMDet | ONNX | float | Qualcomm® QCS8550 (Proxy) | 13.593 ms | 0 - 55 MB | NPU
49
- | RTMDet | ONNX | float | Qualcomm® QCS9075 | 23.627 ms | 5 - 12 MB | NPU
50
- | RTMDet | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 8.307 ms | 3 - 184 MB | NPU
51
- | RTMDet | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 5.973 ms | 5 - 189 MB | NPU
52
- | RTMDet | ONNX | w8a16_mixed_fp16 | Snapdragon® X2 Elite | 11.226 ms | 32 - 32 MB | NPU
53
- | RTMDet | ONNX | w8a16_mixed_fp16 | Snapdragon® X Elite | 29.616 ms | 29 - 29 MB | NPU
54
- | RTMDet | ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Gen 3 Mobile | 22.087 ms | 3 - 383 MB | NPU
55
- | RTMDet | ONNX | w8a16_mixed_fp16 | Qualcomm® QCS8550 (Proxy) | 28.159 ms | 0 - 36 MB | NPU
56
- | RTMDet | ONNX | w8a16_mixed_fp16 | Qualcomm® QCS9075 | 32.76 ms | 2 - 5 MB | NPU
57
- | RTMDet | ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Elite For Galaxy Mobile | 14.374 ms | 1 - 311 MB | NPU
58
- | RTMDet | ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Elite Gen 5 Mobile | 10.413 ms | 3 - 328 MB | NPU
59
- | RTMDet | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 11.722 ms | 0 - 280 MB | NPU
60
- | RTMDet | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 84.037 ms | 0 - 207 MB | NPU
61
- | RTMDet | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 16.034 ms | 0 - 3 MB | NPU
62
- | RTMDet | TFLITE | float | Qualcomm® SA8775P | 23.026 ms | 0 - 209 MB | NPU
63
- | RTMDet | TFLITE | float | Qualcomm® QCS9075 | 25.322 ms | 0 - 62 MB | NPU
64
- | RTMDet | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 38.082 ms | 0 - 347 MB | NPU
65
- | RTMDet | TFLITE | float | Qualcomm® SA7255P | 84.037 ms | 0 - 207 MB | NPU
66
- | RTMDet | TFLITE | float | Qualcomm® SA8295P | 29.911 ms | 0 - 268 MB | NPU
67
- | RTMDet | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 9.169 ms | 0 - 209 MB | NPU
68
- | RTMDet | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 6.836 ms | 0 - 210 MB | NPU
69
 
70
  ## License
71
  * The license for the original implementation of RTMDet can be found
 
15
  RTMDet is a highly efficient model for real-time object detection,capable of predicting both the bounding boxes and classes of objects within an image.It is highly optimized for real-time applications, making it reliable for industrial and commercial use
16
 
17
  This is based on the implementation of RTMDet found [here](https://github.com/open-mmlab/mmdetection/tree/3.x/configs/rtmdet).
18
+ 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/tree/v0.49.1/qai_hub_models/models/rtmdet) 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
  Due to licensing restrictions, we cannot distribute pre-exported model assets for this model.
24
+ Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/tree/v0.49.1/qai_hub_models/models/rtmdet) Python library to compile and export the model with your own:
25
  - Custom weights (e.g., fine-tuned checkpoints)
26
  - Custom input shapes
27
  - Target device and runtime configurations
28
 
29
+ See our repository for [RTMDet on GitHub](https://github.com/qualcomm/ai-hub-models/tree/v0.49.1/qai_hub_models/models/rtmdet) for usage instructions.
30
 
31
 
32
  ## Model Details
 
42
  ## Performance Summary
43
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
44
  |---|---|---|---|---|---|---
45
+ | RTMDet | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 5.977 ms | 5 - 189 MB | NPU
46
+ | RTMDet | ONNX | float | Snapdragon® X2 Elite | 8.176 ms | 53 - 53 MB | NPU
47
+ | RTMDet | ONNX | float | Snapdragon® X Elite | 14.152 ms | 51 - 51 MB | NPU
48
+ | RTMDet | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 10.735 ms | 5 - 236 MB | NPU
49
+ | RTMDet | ONNX | float | Qualcomm® QCS8550 (Proxy) | 13.618 ms | 0 - 54 MB | NPU
50
+ | RTMDet | ONNX | float | Qualcomm® QCS9075 | 23.529 ms | 5 - 12 MB | NPU
51
+ | RTMDet | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 8.301 ms | 3 - 185 MB | NPU
52
+ | RTMDet | ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Elite Gen 5 Mobile | 10.395 ms | 3 - 328 MB | NPU
53
+ | RTMDet | ONNX | w8a16_mixed_fp16 | Snapdragon® X2 Elite | 11.244 ms | 32 - 32 MB | NPU
54
+ | RTMDet | ONNX | w8a16_mixed_fp16 | Snapdragon® X Elite | 29.67 ms | 29 - 29 MB | NPU
55
+ | RTMDet | ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Gen 3 Mobile | 22.111 ms | 3 - 386 MB | NPU
56
+ | RTMDet | ONNX | w8a16_mixed_fp16 | Qualcomm® QCS8550 (Proxy) | 28.29 ms | 2 - 39 MB | NPU
57
+ | RTMDet | ONNX | w8a16_mixed_fp16 | Qualcomm® QCS9075 | 33.104 ms | 2 - 5 MB | NPU
58
+ | RTMDet | ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Elite For Galaxy Mobile | 14.303 ms | 1 - 300 MB | NPU
59
+ | RTMDet | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 6.82 ms | 0 - 210 MB | NPU
60
+ | RTMDet | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 11.725 ms | 0 - 286 MB | NPU
61
+ | RTMDet | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 83.986 ms | 0 - 207 MB | NPU
62
+ | RTMDet | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 15.862 ms | 0 - 3 MB | NPU
63
+ | RTMDet | TFLITE | float | Qualcomm® SA8775P | 22.952 ms | 0 - 208 MB | NPU
64
+ | RTMDet | TFLITE | float | Qualcomm® QCS9075 | 24.387 ms | 0 - 62 MB | NPU
65
+ | RTMDet | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 37.607 ms | 0 - 347 MB | NPU
66
+ | RTMDet | TFLITE | float | Qualcomm® SA7255P | 83.986 ms | 0 - 207 MB | NPU
67
+ | RTMDet | TFLITE | float | Qualcomm® SA8295P | 29.928 ms | 0 - 268 MB | NPU
68
+ | RTMDet | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 9.146 ms | 0 - 208 MB | NPU
69
 
70
  ## License
71
  * The license for the original implementation of RTMDet can be found