---
title: DETR-ResNet50-DC5
canonical_url: "https://www.modelscope.cn/models/Qualcomm-AI-Hub/DETR-ResNet50-DC5"
md_url: "https://www.modelscope.cn/models/Qualcomm-AI-Hub/DETR-ResNet50-DC5.md"
repository: Qualcomm-AI-Hub/DETR-ResNet50-DC5
last_updated: 2026-10-16
license: apache-2.0
downloads: 0
stars: 0
tags:
  - bu_auto
  - android
---

# DETR-ResNet50-DC5

> DETR-ResNet50-DC5 - Qualcomm-AI-Hub 在 ModelScope 开源的模型。DETR-ResNet50-DC5: Optimized for Qualcomm Devices

Qualcomm-AI-Hub/DETR-ResNet50-DC5 是 ModelScope 魔搭社区上的机器学习模型，采用 apache-2.0 许可。

- **Repository**: Qualcomm-AI-Hub/DETR-ResNet50-DC5
- **License**: apache-2.0
- **Tags**: bu_auto, android
- **Downloads**: 0
- **Stars**: 0
- **Last updated**: 2026-10-16

Source: https://www.modelscope.cn/models/Qualcomm-AI-Hub/DETR-ResNet50-DC5

---

![](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/detr_resnet50_dc5/web-assets/model_demo.png)

# DETR-ResNet50-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-ResNet50-DC5 found [here](https://github.com/facebookresearch/detr).
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-cn/blob/main/src/qai_hub_models/models/detr_resnet50_dc5) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).

Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.cn) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) 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](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/detr_resnet50_dc5/releases/v0.62.1/detr_resnet50_dc5-onnx-float.zip)
| ONNX | w8a16_mixed_int16 | Universal | QAIRT 2.45, ONNX Runtime 1.27.1 | [Download](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/detr_resnet50_dc5/releases/v0.62.1/detr_resnet50_dc5-onnx-w8a16_mixed_int16.zip)
| QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/detr_resnet50_dc5/releases/v0.62.1/detr_resnet50_dc5-qnn_dlc-float.zip)
| QNN_DLC | w8a16_mixed_int16 | Universal | QAIRT 2.45 | [Download](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/detr_resnet50_dc5/releases/v0.62.1/detr_resnet50_dc5-qnn_dlc-w8a16_mixed_int16.zip)
| TFLITE | float | Universal | QAIRT 2.45 | [Download](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/detr_resnet50_dc5/releases/v0.62.1/detr_resnet50_dc5-tflite-float.zip)

For more device-specific assets and performance metrics, visit **[DETR-ResNet50-DC5 on Qualcomm® AI Hub](https://aihub.qualcomm.cn/models/detr_resnet50_dc5)**.


### Option 2: Export with Custom Configurations

Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models-cn/blob/main/src/qai_hub_models/models/detr_resnet50_dc5) 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-ResNet50-DC5 on GitHub](https://github.com/qualcomm/ai-hub-models-cn/blob/main/src/qai_hub_models/models/detr_resnet50_dc5) for usage instructions.

## Model Details

**Model Type:** Model_use_case.object_detection

**Model Stats:**
- Input resolution: 480x480
- Model checkpoint: ResNet50-DC5
- Model size (float): 160 MB

## Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|---|---|---|---|---|---|---
| DETR-ResNet50-DC5 | ONNX | float | Snapdragon® X2 Elite | 19.887 ms | 5 - 5 MB | NPU
| DETR-ResNet50-DC5 | ONNX | float | Snapdragon® X Elite | 44.666 ms | 78 - 78 MB | NPU
| DETR-ResNet50-DC5 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 31.647 ms | 3 - 488 MB | NPU
| DETR-ResNet50-DC5 | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 68.417 ms | 5 - 418 MB | NPU
| DETR-ResNet50-DC5 | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 59.183 ms | 5 - 13 MB | NPU
| DETR-ResNet50-DC5 | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 43.257 ms | 0 - 94 MB | NPU
| DETR-ResNet50-DC5 | ONNX | float | Qualcomm® QCS8450 | 68.417 ms | 5 - 418 MB | NPU
| DETR-ResNet50-DC5 | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 57.769 ms | 4 - 12 MB | NPU
| DETR-ResNet50-DC5 | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 44.666 ms | 78 - 78 MB | NPU
| DETR-ResNet50-DC5 | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 22.94 ms | 3 - 375 MB | NPU
| DETR-ResNet50-DC5 | ONNX | float | Snapdragon® 8 Elite Mobile | 22.94 ms | 3 - 375 MB | NPU
| DETR-ResNet50-DC5 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 18.461 ms | 3 - 387 MB | NPU
| DETR-ResNet50-DC5 | ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Gen 3 Mobile | 136.194 ms | 51 - 875 MB | NPU
| DETR-ResNet50-DC5 | ONNX | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ IQ-8275 | 190.978 ms | 51 - 54 MB | NPU
| DETR-ResNet50-DC5 | ONNX | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 171.155 ms | 16 - 80 MB | NPU
| DETR-ResNet50-DC5 | ONNX | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ IQ-9075 | 158.079 ms | 54 - 56 MB | NPU
| DETR-ResNet50-DC5 | ONNX | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ Q-8750 | 113.898 ms | 30 - 550 MB | NPU
| DETR-ResNet50-DC5 | ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Elite Mobile | 113.898 ms | 30 - 550 MB | NPU
| DETR-ResNet50-DC5 | ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Elite Gen 5 Mobile | 99.167 ms | 28 - 755 MB | NPU
| DETR-ResNet50-DC5 | ONNX | w8a16_mixed_int16 | Snapdragon® 8 Gen 3 Mobile | 152.128 ms | 48 - 551 MB | NPU
| DETR-ResNet50-DC5 | ONNX | w8a16_mixed_int16 | Qualcomm® Dragonwing™ Q-8750 | 140.17 ms | 54 - 402 MB | NPU
| DETR-ResNet50-DC5 | ONNX | w8a16_mixed_int16 | Snapdragon® 8 Elite Mobile | 140.17 ms | 54 - 402 MB | NPU
| DETR-ResNet50-DC5 | ONNX | w8a16_mixed_int16 | Snapdragon® 8 Elite Gen 5 Mobile | 111.689 ms | 52 - 438 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | float | Snapdragon® X2 Elite | 21.242 ms | 5 - 5 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | float | Snapdragon® X Elite | 46.618 ms | 5 - 5 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 32.479 ms | 5 - 467 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 71.658 ms | 5 - 397 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 59.531 ms | 5 - 12 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 44.587 ms | 5 - 7 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | float | Qualcomm® SA8775P | 57.771 ms | 0 - 371 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | float | Qualcomm® SA8650P | 57.771 ms | 0 - 371 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | float | Qualcomm® SA8255P | 57.771 ms | 0 - 371 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | float | Qualcomm® QCS8450 | 71.658 ms | 5 - 397 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 58.193 ms | 7 - 13 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 46.618 ms | 5 - 5 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 23.735 ms | 0 - 394 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | float | Qualcomm® SA7255P | 173.078 ms | 3 - 373 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | float | Qualcomm® SA8295P | 64.055 ms | 0 - 305 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 23.735 ms | 0 - 394 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 19.428 ms | 5 - 394 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_fp16 | Snapdragon® X2 Elite | 65.808 ms | 2 - 2 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_fp16 | Snapdragon® X Elite | 105.086 ms | 2 - 2 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_fp16 | Snapdragon® 8 Gen 3 Mobile | 98.134 ms | 2 - 900 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ IQ-8275 | 127.508 ms | 2 - 7 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 113.54 ms | 3 - 5 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® SA8775P | 118.466 ms | 1 - 778 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® SA8650P | 118.466 ms | 1 - 778 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® SA8255P | 118.466 ms | 1 - 778 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ IQ-9075 | 141.538 ms | 2 - 6 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ IQ-X7181 | 105.086 ms | 2 - 2 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ Q-8750 | 79.185 ms | 2 - 782 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® SA7255P | 239.964 ms | 2 - 757 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_fp16 | Snapdragon® 8 Elite Mobile | 79.185 ms | 2 - 782 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_fp16 | Snapdragon® 8 Elite Gen 5 Mobile | 69.835 ms | 2 - 691 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_int16 | Snapdragon® X2 Elite | 22.123 ms | 2 - 2 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_int16 | Snapdragon® X Elite | 42.064 ms | 2 - 2 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_int16 | Snapdragon® 8 Gen 3 Mobile | 31.43 ms | 2 - 732 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® Dragonwing™ IQ-8275 | 39.54 ms | 2 - 7 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 40.431 ms | 2 - 5 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® SA8775P | 41.823 ms | 0 - 610 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® SA8650P | 41.823 ms | 0 - 610 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® SA8255P | 41.823 ms | 0 - 610 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® Dragonwing™ IQ-9075 | 42.413 ms | 1 - 5 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® Dragonwing™ IQ-X7181 | 42.064 ms | 2 - 2 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® Dragonwing™ Q-6690 | 678.622 ms | 2 - 688 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® Dragonwing™ Q-7790 | 61.328 ms | 2 - 526 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® Dragonwing™ Q-8750 | 28.488 ms | 2 - 601 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® SA7255P | 92.132 ms | 1 - 612 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_int16 | Snapdragon® 8 Elite Mobile | 28.488 ms | 2 - 601 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_int16 | Snapdragon® 8 Elite Gen 5 Mobile | 19.443 ms | 2 - 617 MB | NPU
| DETR-ResNet50-DC5 | QNN_DLC | w8a16_mixed_int16 | Snapdragon® 7 Gen 4 Mobile | 61.328 ms | 2 - 526 MB | NPU
| DETR-ResNet50-DC5 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 32.429 ms | 0 - 494 MB | NPU
| DETR-ResNet50-DC5 | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 71.109 ms | 0 - 413 MB | NPU
| DETR-ResNet50-DC5 | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 59.339 ms | 0 - 90 MB | NPU
| DETR-ResNet50-DC5 | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 44.213 ms | 0 - 3 MB | NPU
| DETR-ResNet50-DC5 | TFLITE | float | Qualcomm® SA8775P | 57.499 ms | 0 - 393 MB | NPU
| DETR-ResNet50-DC5 | TFLITE | float | Qualcomm® SA8650P | 57.499 ms | 0 - 393 MB | NPU
| DETR-ResNet50-DC5 | TFLITE | float | Qualcomm® SA8255P | 57.499 ms | 0 - 393 MB | NPU
| DETR-ResNet50-DC5 | TFLITE | float | Qualcomm® QCS8450 | 71.109 ms | 0 - 413 MB | NPU
| DETR-ResNet50-DC5 | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 57.789 ms | 0 - 89 MB | NPU
| DETR-ResNet50-DC5 | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 23.706 ms | 0 - 405 MB | NPU
| DETR-ResNet50-DC5 | TFLITE | float | Qualcomm® SA7255P | 172.479 ms | 0 - 393 MB | NPU
| DETR-ResNet50-DC5 | TFLITE | float | Qualcomm® SA8295P | 63.63 ms | 0 - 327 MB | NPU
| DETR-ResNet50-DC5 | TFLITE | float | Snapdragon® 8 Elite Mobile | 23.706 ms | 0 - 405 MB | NPU
| DETR-ResNet50-DC5 | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 19.223 ms | 0 - 415 MB | NPU

## License
* The license for the original implementation of DETR-ResNet50-DC5 can be found
  [here](https://github.com/facebookresearch/detr/blob/main/LICENSE).

## References
* [End-to-End Object Detection with Transformers](https://arxiv.org/abs/2005.12872)
* [Source Model Implementation](https://github.com/facebookresearch/detr)

## Community
* For questions or feedback, chat with our engineers in [Qualcomm® AI Hub Workbench](https://workbench.aihub.qualcomm.cn).