--- library_name: pytorch license: other tags: - android pipeline_tag: object-detection --- ![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet34_ssd1200/web-assets/model_demo.png) # ResNet34-SSD: Optimized for Qualcomm Devices ResNet34-SSD is a single-stage object detection model that integrates the ResNet34 backbone with the SSD (Single Shot MultiBox Detector) framework. It is optimized for real-time detection tasks and supports multiple deployment backends including PyTorch, TensorFlow, and ONNX. This is based on the implementation of ResNet34-SSD found [here](https://github.com/mlcommons/inference/tree/33894a19c4af6207f7cfdda75f84570f04836de5/vision/classification_and_detection). 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/v0.59.0/src/qai_hub_models/models/resnet34_ssd1200) 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.com) 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://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet34_ssd1200/releases/v0.59.0/resnet34_ssd1200-onnx-float.zip) | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet34_ssd1200/releases/v0.59.0/resnet34_ssd1200-qnn_dlc-float.zip) | TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet34_ssd1200/releases/v0.59.0/resnet34_ssd1200-tflite-float.zip) For more device-specific assets and performance metrics, visit **[ResNet34-SSD on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/resnet34_ssd1200)**. ### Option 2: Export with Custom Configurations Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.59.0/src/qai_hub_models/models/resnet34_ssd1200) 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 [ResNet34-SSD on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.59.0/src/qai_hub_models/models/resnet34_ssd1200) for usage instructions. ## Model Details **Model Type:** Model_use_case.object_detection **Model Stats:** - Model checkpoint: resnet34-ssd1200 - Input resolution: 1x3x1200x1200 - Number of parameters: 20.0M - Model size (float): 76.2 MB ## Performance Summary | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |---|---|---|---|---|---|--- | ResNet34-SSD | ONNX | float | Snapdragon® X2 Elite | 43.332 ms | 17 - 17 MB | NPU | ResNet34-SSD | ONNX | float | Snapdragon® X Elite | 88.328 ms | 30 - 30 MB | NPU | ResNet34-SSD | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 63.01 ms | 3 - 501 MB | NPU | ResNet34-SSD | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 179.687 ms | 17 - 445 MB | NPU | ResNet34-SSD | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 87.363 ms | 0 - 31 MB | NPU | ResNet34-SSD | ONNX | float | Qualcomm® QCS8450 | 179.687 ms | 17 - 445 MB | NPU | ResNet34-SSD | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 133.727 ms | 16 - 36 MB | NPU | ResNet34-SSD | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 88.328 ms | 30 - 30 MB | NPU | ResNet34-SSD | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 52.078 ms | 1 - 423 MB | NPU | ResNet34-SSD | ONNX | float | Snapdragon® 8 Elite Mobile | 52.078 ms | 1 - 423 MB | NPU | ResNet34-SSD | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 38.973 ms | 0 - 489 MB | NPU | ResNet34-SSD | QNN_DLC | float | Snapdragon® X2 Elite | 61.973 ms | 17 - 17 MB | NPU | ResNet34-SSD | QNN_DLC | float | Snapdragon® X Elite | 130.901 ms | 17 - 17 MB | NPU | ResNet34-SSD | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 84.929 ms | 15 - 604 MB | NPU | ResNet34-SSD | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 261.455 ms | 3 - 509 MB | NPU | ResNet34-SSD | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8275 | 481.745 ms | 16 - 384 MB | NPU | ResNet34-SSD | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 129.444 ms | 17 - 19 MB | NPU | ResNet34-SSD | QNN_DLC | float | Qualcomm® SA8775P | 173.627 ms | 16 - 385 MB | NPU | ResNet34-SSD | QNN_DLC | float | Qualcomm® SA8650P | 173.627 ms | 16 - 385 MB | NPU | ResNet34-SSD | QNN_DLC | float | Qualcomm® SA8255P | 173.627 ms | 16 - 385 MB | NPU | ResNet34-SSD | QNN_DLC | float | Qualcomm® QCS8450 | 261.455 ms | 3 - 509 MB | NPU | ResNet34-SSD | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 172.622 ms | 18 - 37 MB | NPU | ResNet34-SSD | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 130.901 ms | 17 - 17 MB | NPU | ResNet34-SSD | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 67.015 ms | 16 - 391 MB | NPU | ResNet34-SSD | QNN_DLC | float | Qualcomm® SA7255P | 481.745 ms | 16 - 384 MB | NPU | ResNet34-SSD | QNN_DLC | float | Qualcomm® SA8295P | 183.702 ms | 0 - 329 MB | NPU | ResNet34-SSD | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 67.015 ms | 16 - 391 MB | NPU | ResNet34-SSD | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 52.389 ms | 8 - 547 MB | NPU | ResNet34-SSD | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 107.47 ms | 0 - 542 MB | NPU | ResNet34-SSD | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 233.539 ms | 1 - 621 MB | NPU | ResNet34-SSD | TFLITE | float | Qualcomm® Dragonwing™ QCS8275 | 513.176 ms | 0 - 378 MB | NPU | ResNet34-SSD | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 146.676 ms | 0 - 3 MB | NPU | ResNet34-SSD | TFLITE | float | Qualcomm® SA8775P | 183.849 ms | 1 - 427 MB | NPU | ResNet34-SSD | TFLITE | float | Qualcomm® SA8650P | 183.849 ms | 1 - 427 MB | NPU | ResNet34-SSD | TFLITE | float | Qualcomm® SA8255P | 183.849 ms | 1 - 427 MB | NPU | ResNet34-SSD | TFLITE | float | Qualcomm® QCS8450 | 233.539 ms | 1 - 621 MB | NPU | ResNet34-SSD | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 181.406 ms | 0 - 65 MB | NPU | ResNet34-SSD | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 87.816 ms | 0 - 403 MB | NPU | ResNet34-SSD | TFLITE | float | Qualcomm® SA7255P | 513.176 ms | 0 - 378 MB | NPU | ResNet34-SSD | TFLITE | float | Qualcomm® SA8295P | 202.221 ms | 0 - 353 MB | NPU | ResNet34-SSD | TFLITE | float | Snapdragon® 8 Elite Mobile | 87.816 ms | 0 - 403 MB | NPU | ResNet34-SSD | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 72.472 ms | 0 - 570 MB | NPU ## License * The license for the original implementation of ResNet34-SSD can be found [here](https://github.com/mlcommons/inference/blob/33894a19c4af6207f7cfdda75f84570f04836de5/LICENSE.md). ## References * [Source Model Implementation](https://github.com/mlcommons/inference/tree/33894a19c4af6207f7cfdda75f84570f04836de5/vision/classification_and_detection) ## Community * Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI. * For questions or feedback please [reach out to us](mailto:ai-hub-support@qti.qualcomm.com).