Efficient and Robust 2D-to-BEV Representation Learning via Geometry-guided Kernel Transformer
Paper
•
2206.04584
•
Published
GKT: Optimized for Qualcomm Devices
Geometry-guided Kernel Transformer is a machine learning model for generating a birds eye view represenation from the sensors(cameras) mounted on a vehicle.
This is based on the implementation of GKT found [here](https://github.com/hustvl/GKT/ https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/Qualcomm+AI+Hub+Proprietary+License.pdf). 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 |
|---|---|---|---|---|
| PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | QAIRT 2.37, ONNX Runtime 1.23.0 | Download |
| PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | QAIRT 2.37, ONNX Runtime 1.23.0 | Download |
| PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS8550 (Proxy) | QAIRT 2.37, ONNX Runtime 1.23.0 | Download |
| PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | QAIRT 2.37, ONNX Runtime 1.23.0 | Download |
| PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | QAIRT 2.37, ONNX Runtime 1.23.0 | Download |
| PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS9075 | QAIRT 2.37, ONNX Runtime 1.23.0 | Download |
| PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Snapdragon® X Elite | QAIRT 2.37, ONNX Runtime 1.23.0 | Download |
| PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Gen 3 Mobile | QAIRT 2.37, ONNX Runtime 1.23.0 | Download |
| PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Qualcomm® QCS8550 (Proxy) | QAIRT 2.37, ONNX Runtime 1.23.0 | Download |
| PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Elite For Galaxy Mobile | QAIRT 2.37, ONNX Runtime 1.23.0 | Download |
| PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Elite Gen 5 Mobile | QAIRT 2.37, ONNX Runtime 1.23.0 | Download |
| PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Qualcomm® QCS9075 | QAIRT 2.37, ONNX Runtime 1.23.0 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | QAIRT 2.42 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | QAIRT 2.42 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® QCS8275 (Proxy) | QAIRT 2.42 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® QCS8550 (Proxy) | QAIRT 2.42 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® SA8775P | QAIRT 2.42 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite For Galaxy Mobile | QAIRT 2.42 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | QAIRT 2.42 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® SA7255P | QAIRT 2.42 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® SA8295P | QAIRT 2.42 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® QCS9075 | QAIRT 2.42 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® QCS8450 (Proxy) | QAIRT 2.42 | Download |
| QNN_CONTEXT_BINARY | w8a16_mixed_fp16 | Snapdragon® X Elite | QAIRT 2.42 | Download |
| QNN_CONTEXT_BINARY | w8a16_mixed_fp16 | Snapdragon® 8 Gen 3 Mobile | QAIRT 2.42 | Download |
| QNN_CONTEXT_BINARY | w8a16_mixed_fp16 | Qualcomm® QCS8275 (Proxy) | QAIRT 2.42 | Download |
| QNN_CONTEXT_BINARY | w8a16_mixed_fp16 | Qualcomm® QCS8550 (Proxy) | QAIRT 2.42 | Download |
| QNN_CONTEXT_BINARY | w8a16_mixed_fp16 | Qualcomm® SA8775P | QAIRT 2.42 | Download |
| QNN_CONTEXT_BINARY | w8a16_mixed_fp16 | Snapdragon® 8 Elite For Galaxy Mobile | QAIRT 2.42 | Download |
| QNN_CONTEXT_BINARY | w8a16_mixed_fp16 | Snapdragon® 8 Elite Gen 5 Mobile | QAIRT 2.42 | Download |
| QNN_CONTEXT_BINARY | w8a16_mixed_fp16 | Qualcomm® SA7255P | QAIRT 2.42 | Download |
| QNN_CONTEXT_BINARY | w8a16_mixed_fp16 | Qualcomm® SA8295P | QAIRT 2.42 | Download |
| QNN_CONTEXT_BINARY | w8a16_mixed_fp16 | Qualcomm® QCS9075 | QAIRT 2.42 | Download |
| QNN_CONTEXT_BINARY | w8a16_mixed_fp16 | Qualcomm® QCS8450 (Proxy) | QAIRT 2.42 | Download |
For more device-specific assets and performance metrics, visit GKT 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 GKT on GitHub for usage instructions.
Model Details
Model Type: Model_use_case.driver_assistance
Model Stats:
- Model checkpoint: map_segmentation_gkt_7x1_conv_setting2.ckpt
- Input resolution: 1 x 6 x 3 x 224 x 480
- Number of parameters: 1.18M
- Model size: 4.66 MB
Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |
|---|---|---|---|---|---|---|
| GKT | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | 77.549 ms | 7 - 7 MB | NPU |
| GKT | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | 56.522 ms | 8 - 15 MB | NPU |
| GKT | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS8550 (Proxy) | 81.44 ms | 8 - 19 MB | NPU |
| GKT | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS9075 | 83.044 ms | 7 - 10 MB | NPU |
| GKT | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 51.556 ms | 2 - 10 MB | NPU |
| GKT | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 38.135 ms | 8 - 18 MB | NPU |
| GKT | PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Snapdragon® X Elite | 130.45 ms | 7 - 7 MB | NPU |
| GKT | PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Gen 3 Mobile | 95.525 ms | 6 - 14 MB | NPU |
| GKT | PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Qualcomm® QCS8550 (Proxy) | 129.902 ms | 1 - 10 MB | NPU |
| GKT | PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Qualcomm® QCS9075 | 132.041 ms | 4 - 6 MB | NPU |
| GKT | PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Elite For Galaxy Mobile | 83.007 ms | 0 - 12 MB | NPU |
| GKT | PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Elite Gen 5 Mobile | 71.133 ms | 4 - 15 MB | NPU |
| GKT | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | 105.686 ms | 7 - 7 MB | NPU |
| GKT | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | 77.938 ms | 8 - 15 MB | NPU |
| GKT | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8275 (Proxy) | 188.21 ms | 0 - 8 MB | NPU |
| GKT | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8550 (Proxy) | 110.633 ms | 8 - 9 MB | NPU |
| GKT | QNN_CONTEXT_BINARY | float | Qualcomm® SA8775P | 110.755 ms | 0 - 6 MB | NPU |
| GKT | QNN_CONTEXT_BINARY | float | Qualcomm® QCS9075 | 109.963 ms | 7 - 17 MB | NPU |
| GKT | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8450 (Proxy) | 204.587 ms | 8 - 16 MB | NPU |
| GKT | QNN_CONTEXT_BINARY | float | Qualcomm® SA7255P | 188.21 ms | 0 - 8 MB | NPU |
| GKT | QNN_CONTEXT_BINARY | float | Qualcomm® SA8295P | 140.428 ms | 0 - 6 MB | NPU |
| GKT | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite For Galaxy Mobile | 71.057 ms | 0 - 9 MB | NPU |
| GKT | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | 57.265 ms | 7 - 17 MB | NPU |
| GKT | QNN_CONTEXT_BINARY | w8a16_mixed_fp16 | Snapdragon® X Elite | 128.698 ms | 4 - 4 MB | NPU |
| GKT | QNN_CONTEXT_BINARY | w8a16_mixed_fp16 | Snapdragon® 8 Gen 3 Mobile | 94.5 ms | 4 - 11 MB | NPU |
| GKT | QNN_CONTEXT_BINARY | w8a16_mixed_fp16 | Qualcomm® QCS8275 (Proxy) | 192.408 ms | 1 - 8 MB | NPU |
| GKT | QNN_CONTEXT_BINARY | w8a16_mixed_fp16 | Qualcomm® QCS8550 (Proxy) | 127.219 ms | 5 - 6 MB | NPU |
| GKT | QNN_CONTEXT_BINARY | w8a16_mixed_fp16 | Qualcomm® SA8775P | 129.389 ms | 0 - 6 MB | NPU |
| GKT | QNN_CONTEXT_BINARY | w8a16_mixed_fp16 | Qualcomm® QCS9075 | 129.942 ms | 2 - 7 MB | NPU |
| GKT | QNN_CONTEXT_BINARY | w8a16_mixed_fp16 | Qualcomm® QCS8450 (Proxy) | 138.637 ms | 4 - 13 MB | NPU |
| GKT | QNN_CONTEXT_BINARY | w8a16_mixed_fp16 | Qualcomm® SA7255P | 192.408 ms | 1 - 8 MB | NPU |
| GKT | QNN_CONTEXT_BINARY | w8a16_mixed_fp16 | Qualcomm® SA8295P | 153.11 ms | 0 - 6 MB | NPU |
| GKT | QNN_CONTEXT_BINARY | w8a16_mixed_fp16 | Snapdragon® 8 Elite For Galaxy Mobile | 80.584 ms | 0 - 13 MB | NPU |
| GKT | QNN_CONTEXT_BINARY | w8a16_mixed_fp16 | Snapdragon® 8 Elite Gen 5 Mobile | 69.942 ms | 4 - 13 MB | NPU |
License
- The license for the original implementation of GKT can be found here.
References
- Efficient and Robust 2D-to-BEV Representation Learning via Geometry-guided Kernel Transformer
- [Source Model Implementation](https://github.com/hustvl/GKT/ https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/Qualcomm+AI+Hub+Proprietary+License.pdf)
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.
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support