SalsaNext: Optimized for Qualcomm Devices
SalsaNext is a LiDAR-based model designed for efficient and accurate semantic segmentation.
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 2.42, ONNX Runtime 1.25.0 | 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 SalsaNext 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 SalsaNext on GitHub for usage instructions.
Model Details
Model Type: Model_use_case.semantic_segmentation
Model Stats:
- Model checkpoint: SalsaNext
- Input resolution: 1x5x64x2048
- Number of parameters: 6.71M
- Model size (float): 25.7 MB
Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |
|---|---|---|---|---|---|---|
| SalsaNext | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 15.634 ms | 24 - 215 MB | NPU |
| SalsaNext | ONNX | float | Snapdragon® X2 Elite | 14.887 ms | 210 - 210 MB | NPU |
| SalsaNext | ONNX | float | Snapdragon® X Elite | 32.382 ms | 157 - 157 MB | NPU |
| SalsaNext | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 25.806 ms | 25 - 292 MB | NPU |
| SalsaNext | ONNX | float | Qualcomm® QCS8550 (Proxy) | 34.185 ms | 0 - 44 MB | NPU |
| SalsaNext | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 18.437 ms | 24 - 211 MB | NPU |
| SalsaNext | ONNX | float | Qualcomm® QCS9075 | 39.954 ms | 26 - 72 MB | NPU |
| SalsaNext | ONNX | float | Qualcomm® QCS8750 | 18.437 ms | 24 - 211 MB | NPU |
| SalsaNext | ONNX | float | Qualcomm® QCS7181 | 32.382 ms | 157 - 157 MB | NPU |
| SalsaNext | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 13.206 ms | 3 - 291 MB | NPU |
| SalsaNext | QNN_DLC | float | Snapdragon® X2 Elite | 13.8 ms | 3 - 3 MB | NPU |
| SalsaNext | QNN_DLC | float | Snapdragon® X Elite | 32.078 ms | 3 - 3 MB | NPU |
| SalsaNext | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 25.313 ms | 0 - 330 MB | NPU |
| SalsaNext | QNN_DLC | float | Qualcomm® QCS8275 | 118.614 ms | 2 - 248 MB | NPU |
| SalsaNext | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 33.136 ms | 3 - 4 MB | NPU |
| SalsaNext | QNN_DLC | float | Qualcomm® SA8775P | 39.296 ms | 1 - 247 MB | NPU |
| SalsaNext | QNN_DLC | float | Qualcomm® SA8650P | 39.296 ms | 1 - 247 MB | NPU |
| SalsaNext | QNN_DLC | float | Qualcomm® SA8255P | 39.296 ms | 1 - 247 MB | NPU |
| SalsaNext | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 65.454 ms | 2 - 347 MB | NPU |
| SalsaNext | QNN_DLC | float | Qualcomm® SA7255P | 118.614 ms | 2 - 248 MB | NPU |
| SalsaNext | QNN_DLC | float | Qualcomm® SA8295P | 51.612 ms | 0 - 259 MB | NPU |
| SalsaNext | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 18.577 ms | 0 - 265 MB | NPU |
| SalsaNext | QNN_DLC | float | Qualcomm® QCS9075 | 37.085 ms | 3 - 17 MB | NPU |
| SalsaNext | QNN_DLC | float | Qualcomm® QCS8750 | 18.577 ms | 0 - 265 MB | NPU |
| SalsaNext | QNN_DLC | float | Qualcomm® QCS7181 | 32.078 ms | 3 - 3 MB | NPU |
| SalsaNext | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 13.216 ms | 10 - 300 MB | NPU |
| SalsaNext | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 25.478 ms | 9 - 345 MB | NPU |
| SalsaNext | TFLITE | float | Qualcomm® QCS8275 | 119.479 ms | 10 - 257 MB | NPU |
| SalsaNext | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 34.016 ms | 10 - 29 MB | NPU |
| SalsaNext | TFLITE | float | Qualcomm® SA8775P | 39.355 ms | 10 - 257 MB | NPU |
| SalsaNext | TFLITE | float | Qualcomm® SA8650P | 39.355 ms | 10 - 257 MB | NPU |
| SalsaNext | TFLITE | float | Qualcomm® SA8255P | 39.355 ms | 10 - 257 MB | NPU |
| SalsaNext | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 70.899 ms | 10 - 362 MB | NPU |
| SalsaNext | TFLITE | float | Qualcomm® SA7255P | 119.479 ms | 10 - 257 MB | NPU |
| SalsaNext | TFLITE | float | Qualcomm® SA8295P | 51.686 ms | 10 - 271 MB | NPU |
| SalsaNext | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 18.41 ms | 9 - 274 MB | NPU |
| SalsaNext | TFLITE | float | Qualcomm® QCS9075 | 37.268 ms | 10 - 39 MB | NPU |
| SalsaNext | TFLITE | float | Qualcomm® QCS8750 | 18.41 ms | 9 - 274 MB | NPU |
License
- The license for the original implementation of SalsaNext can be found here.
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.