RangeNet-Plus-Plus: Optimized for Qualcomm Devices
RangeNet-Plus-Plus (also stylized as RangeNet++) projects a LiDAR point cloud onto a 5-channel range image (depth, x, y, z, intensity) and applies a DarkNet-53 encoder with a decoder head to predict per-point semantic class labels in real time.
This is based on the implementation of RangeNet-Plus-Plus 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 2.50, ONNX Runtime 1.30.0 | Download |
| ONNX | w8a16 | Universal | QAIRT 2.50, ONNX Runtime 1.30.0 | Download |
| QNN_DLC | float | Universal | QAIRT 2.50 | Download |
| QNN_DLC | w8a16 | Universal | QAIRT 2.50 | Download |
| TFLITE | float | Universal | QAIRT 2.50 | Download |
For more device-specific assets and performance metrics, visit RangeNet-Plus-Plus 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 RangeNet-Plus-Plus on GitHub for usage instructions.
Model Details
Model Type: Model_use_case.driver_assistance
Model Stats:
- Backbone: DarkNet-53
- Input channels: 5
- Input resolution: 64x2048
- Model checkpoint: darknet53_rangenet++
- Number of output classes: 20
Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |
|---|---|---|---|---|---|---|
| RangeNet-Plus-Plus | ONNX | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 43.643 ms | 21 - 357 MB | NPU |
| RangeNet-Plus-Plus | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 59.496 ms | 23 - 334 MB | NPU |
| RangeNet-Plus-Plus | ONNX | float | Snapdragon® X2 Elite | 49.351 ms | 35 - 35 MB | NPU |
| RangeNet-Plus-Plus | ONNX | float | Snapdragon® X Elite | 98.493 ms | 102 - 102 MB | NPU |
| RangeNet-Plus-Plus | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 77.022 ms | 0 - 449 MB | NPU |
| RangeNet-Plus-Plus | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 204.893 ms | 0 - 445 MB | NPU |
| RangeNet-Plus-Plus | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 176.809 ms | 21 - 28 MB | NPU |
| RangeNet-Plus-Plus | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 106.323 ms | 23 - 25 MB | NPU |
| RangeNet-Plus-Plus | ONNX | float | Qualcomm® QCS8450 | 204.893 ms | 0 - 445 MB | NPU |
| RangeNet-Plus-Plus | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 151.522 ms | 22 - 28 MB | NPU |
| RangeNet-Plus-Plus | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 98.493 ms | 102 - 102 MB | NPU |
| RangeNet-Plus-Plus | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 59.496 ms | 23 - 334 MB | NPU |
| RangeNet-Plus-Plus | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 27.943 ms | 12 - 353 MB | NPU |
| RangeNet-Plus-Plus | ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 38.489 ms | 11 - 344 MB | NPU |
| RangeNet-Plus-Plus | ONNX | w8a16 | Snapdragon® X2 Elite | 32.559 ms | 18 - 18 MB | NPU |
| RangeNet-Plus-Plus | ONNX | w8a16 | Snapdragon® X Elite | 61.13 ms | 52 - 52 MB | NPU |
| RangeNet-Plus-Plus | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 48.155 ms | 0 - 461 MB | NPU |
| RangeNet-Plus-Plus | ONNX | w8a16 | Snapdragon® 8 Gen 1 Mobile | 98.848 ms | 0 - 569 MB | NPU |
| RangeNet-Plus-Plus | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS6490 | 408.444 ms | 24 - 28 MB | NPU |
| RangeNet-Plus-Plus | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | 62.711 ms | 10 - 16 MB | NPU |
| RangeNet-Plus-Plus | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 63.873 ms | 0 - 57 MB | NPU |
| RangeNet-Plus-Plus | ONNX | w8a16 | Qualcomm® QCS8450 | 98.848 ms | 0 - 569 MB | NPU |
| RangeNet-Plus-Plus | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 58.204 ms | 14 - 18 MB | NPU |
| RangeNet-Plus-Plus | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 61.13 ms | 52 - 52 MB | NPU |
| RangeNet-Plus-Plus | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 1536.611 ms | 0 - 722 MB | NPU |
| RangeNet-Plus-Plus | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 116.805 ms | 29 - 475 MB | NPU |
| RangeNet-Plus-Plus | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 38.489 ms | 11 - 344 MB | NPU |
| RangeNet-Plus-Plus | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 116.805 ms | 29 - 475 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 43.72 ms | 3 - 332 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 60.681 ms | 0 - 306 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | float | Snapdragon® X2 Elite | 49.857 ms | 3 - 3 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | float | Snapdragon® X Elite | 98.523 ms | 3 - 3 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 77.193 ms | 3 - 449 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 175.566 ms | 0 - 436 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 176.667 ms | 3 - 18 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 100.743 ms | 3 - 4 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | float | Qualcomm® SA8775P | 150.255 ms | 1 - 321 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | float | Qualcomm® SA8650P | 150.255 ms | 1 - 321 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | float | Qualcomm® SA8255P | 150.255 ms | 1 - 321 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | float | Qualcomm® QCS8450 | 175.566 ms | 0 - 436 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 150.504 ms | 5 - 19 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 98.523 ms | 3 - 3 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 60.681 ms | 0 - 306 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | float | Qualcomm® SA8295P | 166.785 ms | 0 - 313 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 26.601 ms | 1 - 331 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 41.031 ms | 1 - 326 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 31.891 ms | 1 - 1 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | w8a16 | Snapdragon® X Elite | 61.205 ms | 1 - 1 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 48.459 ms | 1 - 455 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | w8a16 | Snapdragon® 8 Gen 1 Mobile | 96.201 ms | 1 - 563 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ QCS6490 | 403.9 ms | 3 - 11 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | 63.471 ms | 1 - 10 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 61.55 ms | 1 - 465 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | w8a16 | Qualcomm® SA8775P | 58.335 ms | 1 - 318 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | w8a16 | Qualcomm® SA8650P | 58.335 ms | 1 - 318 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | w8a16 | Qualcomm® SA8255P | 58.335 ms | 1 - 318 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | w8a16 | Qualcomm® QCS8450 | 96.201 ms | 1 - 563 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 58.112 ms | 1 - 10 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 61.205 ms | 1 - 1 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 1589.712 ms | 1 - 704 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 116.336 ms | 1 - 436 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 41.031 ms | 1 - 326 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | w8a16 | Qualcomm® SA8295P | 101.827 ms | 2 - 412 MB | NPU |
| RangeNet-Plus-Plus | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 116.336 ms | 1 - 436 MB | NPU |
| RangeNet-Plus-Plus | TFLITE | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 43.112 ms | 12 - 320 MB | NPU |
| RangeNet-Plus-Plus | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 61.24 ms | 12 - 294 MB | NPU |
| RangeNet-Plus-Plus | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 77.744 ms | 11 - 519 MB | NPU |
| RangeNet-Plus-Plus | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 170.829 ms | 13 - 504 MB | NPU |
| RangeNet-Plus-Plus | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 178.898 ms | 13 - 134 MB | NPU |
| RangeNet-Plus-Plus | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 103.55 ms | 13 - 148 MB | NPU |
| RangeNet-Plus-Plus | TFLITE | float | Qualcomm® SA8775P | 152.693 ms | 13 - 308 MB | NPU |
| RangeNet-Plus-Plus | TFLITE | float | Qualcomm® SA8650P | 152.693 ms | 13 - 308 MB | NPU |
| RangeNet-Plus-Plus | TFLITE | float | Qualcomm® SA8255P | 152.693 ms | 13 - 308 MB | NPU |
| RangeNet-Plus-Plus | TFLITE | float | Qualcomm® QCS8450 | 170.829 ms | 13 - 504 MB | NPU |
| RangeNet-Plus-Plus | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 153.757 ms | 12 - 132 MB | NPU |
| RangeNet-Plus-Plus | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 61.24 ms | 12 - 294 MB | NPU |
| RangeNet-Plus-Plus | TFLITE | float | Qualcomm® SA7255P | 2403.316 ms | 11 - 39 MB | GPU |
| RangeNet-Plus-Plus | TFLITE | float | Qualcomm® SA8295P | 167.72 ms | 13 - 301 MB | NPU |
License
- The license for the original implementation of RangeNet-Plus-Plus 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.
