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 2.45, ONNX Runtime 1.25.0 | Download |
| TFLITE | float | Universal | QAIRT 2.45 | 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:
- Model checkpoint: darknet53_rangenet++
- Input resolution: 64x2048
- Input channels: 5
- Number of output classes: 20
- Backbone: DarkNet-53
Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |
|---|---|---|---|---|---|---|
| RangeNet-Plus-Plus | ONNX | float | Snapdragon® X2 Elite | 49.351 ms | 210 - 210 MB | NPU |
| RangeNet-Plus-Plus | ONNX | float | Snapdragon® X Elite | 99.787 ms | 147 - 147 MB | NPU |
| RangeNet-Plus-Plus | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 75.688 ms | 47 - 492 MB | NPU |
| RangeNet-Plus-Plus | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 214.848 ms | 3 - 441 MB | NPU |
| RangeNet-Plus-Plus | ONNX | float | Qualcomm® QCS8550 (Proxy) | 102.175 ms | 0 - 113 MB | NPU |
| RangeNet-Plus-Plus | ONNX | float | Qualcomm® QCS8450 | 214.848 ms | 3 - 441 MB | NPU |
| RangeNet-Plus-Plus | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 42.224 ms | 2 - 342 MB | NPU |
| RangeNet-Plus-Plus | ONNX | float | Qualcomm® QCS9075 | 159.825 ms | 2 - 48 MB | NPU |
| RangeNet-Plus-Plus | ONNX | float | Snapdragon® 8 Elite Mobile | 59.183 ms | 1 - 321 MB | NPU |
| RangeNet-Plus-Plus | ONNX | float | Qualcomm® QCS8750 | 59.183 ms | 1 - 321 MB | NPU |
| RangeNet-Plus-Plus | ONNX | float | Qualcomm® QCS7181 | 99.787 ms | 147 - 147 MB | NPU |
| RangeNet-Plus-Plus | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 78.955 ms | 1 - 510 MB | NPU |
| RangeNet-Plus-Plus | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 197.635 ms | 1 - 499 MB | NPU |
| RangeNet-Plus-Plus | TFLITE | float | Qualcomm® QCS8275 | 596.038 ms | 1 - 308 MB | NPU |
| RangeNet-Plus-Plus | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 106.778 ms | 0 - 96 MB | NPU |
| RangeNet-Plus-Plus | TFLITE | float | Qualcomm® SA8775P | 1229.142 ms | 0 - 29 MB | GPU |
| RangeNet-Plus-Plus | TFLITE | float | Qualcomm® SA8650P | 1229.142 ms | 0 - 29 MB | GPU |
| RangeNet-Plus-Plus | TFLITE | float | Qualcomm® SA8255P | 1229.142 ms | 0 - 29 MB | GPU |
| RangeNet-Plus-Plus | TFLITE | float | Qualcomm® QCS8450 | 197.635 ms | 1 - 499 MB | NPU |
| RangeNet-Plus-Plus | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 43.272 ms | 6 - 323 MB | NPU |
| RangeNet-Plus-Plus | TFLITE | float | Qualcomm® SA7255P | 596.038 ms | 1 - 308 MB | NPU |
| RangeNet-Plus-Plus | TFLITE | float | Qualcomm® QCS9075 | 167.687 ms | 0 - 107 MB | NPU |
| RangeNet-Plus-Plus | TFLITE | float | Snapdragon® 8 Elite Mobile | 60.072 ms | 1 - 294 MB | NPU |
| RangeNet-Plus-Plus | TFLITE | float | Qualcomm® SA8295P | 172.041 ms | 1 - 302 MB | NPU |
| RangeNet-Plus-Plus | TFLITE | float | Qualcomm® QCS8750 | 60.072 ms | 1 - 294 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.