DETR-ResNet101-DC5: Optimized for Qualcomm Devices
DETR is a machine learning model that can detect objects (trained on COCO dataset).
This is based on the implementation of DETR-ResNet101-DC5 found here. This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the Qualcomm® AI Hub Models library to export with custom configurations. More details on model performance across various devices, can be found here.
Qualcomm AI Hub Models uses Qualcomm AI Hub Workbench to compile, profile, and evaluate this model. Sign up to run these models on a hosted Qualcomm® device.
Getting Started
There are two ways to deploy this model on your device:
Option 1: Download Pre-Exported Models
Below are pre-exported model assets ready for deployment.
| Runtime | Precision | Chipset | SDK Versions | Download |
|---|---|---|---|---|
| ONNX | float | Universal | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| QNN_DLC | float | Universal | QAIRT 2.45 | Download |
| TFLITE | float | Universal | QAIRT 2.45 | Download |
For more device-specific assets and performance metrics, visit DETR-ResNet101-DC5 on Qualcomm® AI Hub.
Option 2: Export with Custom Configurations
Use the Qualcomm® AI Hub Models Python library to compile and export the model with your own:
- Custom weights (e.g., fine-tuned checkpoints)
- Custom input shapes
- Target device and runtime configurations
This option is ideal if you need to customize the model beyond the default configuration provided here.
See our repository for DETR-ResNet101-DC5 on GitHub for usage instructions.
Model Details
Model Type: Model_use_case.object_detection
Model Stats:
- Input resolution: 480x480
- Model checkpoint: ResNet101-DC5
- Model size (float): 232 MB
Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |
|---|---|---|---|---|---|---|
| DETR-ResNet101-DC5 | ONNX | float | Snapdragon® X2 Elite | 23.605 ms | 5 - 5 MB | NPU |
| DETR-ResNet101-DC5 | ONNX | float | Snapdragon® X Elite | 51.621 ms | 115 - 115 MB | NPU |
| DETR-ResNet101-DC5 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 36.488 ms | 3 - 568 MB | NPU |
| DETR-ResNet101-DC5 | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 84.44 ms | 1 - 476 MB | NPU |
| DETR-ResNet101-DC5 | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 72.45 ms | 5 - 13 MB | NPU |
| DETR-ResNet101-DC5 | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 49.776 ms | 0 - 127 MB | NPU |
| DETR-ResNet101-DC5 | ONNX | float | Qualcomm® QCS8450 | 84.44 ms | 1 - 476 MB | NPU |
| DETR-ResNet101-DC5 | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 68.967 ms | 5 - 12 MB | NPU |
| DETR-ResNet101-DC5 | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 51.621 ms | 115 - 115 MB | NPU |
| DETR-ResNet101-DC5 | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 27.111 ms | 0 - 434 MB | NPU |
| DETR-ResNet101-DC5 | ONNX | float | Snapdragon® 8 Elite Mobile | 27.111 ms | 0 - 434 MB | NPU |
| DETR-ResNet101-DC5 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 21.759 ms | 1 - 445 MB | NPU |
| DETR-ResNet101-DC5 | QNN_DLC | float | Snapdragon® X2 Elite | 24.822 ms | 5 - 5 MB | NPU |
| DETR-ResNet101-DC5 | QNN_DLC | float | Snapdragon® X Elite | 53.236 ms | 5 - 5 MB | NPU |
| DETR-ResNet101-DC5 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 37.488 ms | 5 - 559 MB | NPU |
| DETR-ResNet101-DC5 | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 111.47 ms | 0 - 465 MB | NPU |
| DETR-ResNet101-DC5 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 73.567 ms | 5 - 12 MB | NPU |
| DETR-ResNet101-DC5 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 220.264 ms | 1 - 427 MB | NPU |
| DETR-ResNet101-DC5 | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 51.313 ms | 5 - 22 MB | NPU |
| DETR-ResNet101-DC5 | QNN_DLC | float | Qualcomm® SA8775P | 69.826 ms | 1 - 427 MB | NPU |
| DETR-ResNet101-DC5 | QNN_DLC | float | Qualcomm® SA8650P | 69.826 ms | 1 - 427 MB | NPU |
| DETR-ResNet101-DC5 | QNN_DLC | float | Qualcomm® SA8255P | 69.826 ms | 1 - 427 MB | NPU |
| DETR-ResNet101-DC5 | QNN_DLC | float | Qualcomm® QCS8450 | 111.47 ms | 0 - 465 MB | NPU |
| DETR-ResNet101-DC5 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 69.702 ms | 7 - 13 MB | NPU |
| DETR-ResNet101-DC5 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 53.236 ms | 5 - 5 MB | NPU |
| DETR-ResNet101-DC5 | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 28.266 ms | 0 - 449 MB | NPU |
| DETR-ResNet101-DC5 | QNN_DLC | float | Qualcomm® SA7255P | 220.264 ms | 1 - 427 MB | NPU |
| DETR-ResNet101-DC5 | QNN_DLC | float | Qualcomm® SA8295P | 86.064 ms | 0 - 341 MB | NPU |
| DETR-ResNet101-DC5 | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 28.266 ms | 0 - 449 MB | NPU |
| DETR-ResNet101-DC5 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 22.655 ms | 5 - 450 MB | NPU |
| DETR-ResNet101-DC5 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 37.347 ms | 0 - 604 MB | NPU |
| DETR-ResNet101-DC5 | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 84.783 ms | 1 - 479 MB | NPU |
| DETR-ResNet101-DC5 | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 73.179 ms | 0 - 127 MB | NPU |
| DETR-ResNet101-DC5 | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 219.468 ms | 0 - 471 MB | NPU |
| DETR-ResNet101-DC5 | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 50.489 ms | 0 - 3 MB | NPU |
| DETR-ResNet101-DC5 | TFLITE | float | Qualcomm® SA8775P | 69.577 ms | 0 - 471 MB | NPU |
| DETR-ResNet101-DC5 | TFLITE | float | Qualcomm® SA8650P | 69.577 ms | 0 - 471 MB | NPU |
| DETR-ResNet101-DC5 | TFLITE | float | Qualcomm® SA8255P | 69.577 ms | 0 - 471 MB | NPU |
| DETR-ResNet101-DC5 | TFLITE | float | Qualcomm® QCS8450 | 84.783 ms | 1 - 479 MB | NPU |
| DETR-ResNet101-DC5 | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 69.774 ms | 0 - 126 MB | NPU |
| DETR-ResNet101-DC5 | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 28.054 ms | 3 - 482 MB | NPU |
| DETR-ResNet101-DC5 | TFLITE | float | Qualcomm® SA7255P | 219.468 ms | 0 - 471 MB | NPU |
| DETR-ResNet101-DC5 | TFLITE | float | Qualcomm® SA8295P | 76.567 ms | 0 - 357 MB | NPU |
| DETR-ResNet101-DC5 | TFLITE | float | Snapdragon® 8 Elite Mobile | 28.054 ms | 3 - 482 MB | NPU |
| DETR-ResNet101-DC5 | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 22.267 ms | 0 - 496 MB | NPU |
License
- The license for the original implementation of DETR-ResNet101-DC5 can be found here.
References
Community
- Join our AI Hub Slack community to collaborate, post questions and learn more about on-device AI.
- For questions or feedback please reach out to us.
