Depth-Anything: Optimized for Mobile Deployment
Deep Convolutional Neural Network model for depth estimation
Depth Anything is designed for estimating depth at each point in an image.
This model is an implementation of Depth-Anything found here.
This repository provides scripts to run Depth-Anything on Qualcomm® devices. More details on model performance across various devices, can be found here.
Model Details
- Model Type: Depth estimation
- Model Stats:
- Model checkpoint: DepthAnything_Small
- Input resolution: 518x518
- Number of parameters: 24.8M
- Model size: 94 MB
Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model |
---|---|---|---|---|---|---|---|---|
Depth-Anything | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 329.103 ms | 0 - 107 MB | FP16 | NPU | Depth-Anything.tflite |
Depth-Anything | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 367.114 ms | 3 - 56 MB | FP16 | NPU | Depth-Anything.so |
Depth-Anything | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 231.904 ms | 0 - 63 MB | FP16 | NPU | Depth-Anything.onnx |
Depth-Anything | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 251.12 ms | 0 - 249 MB | FP16 | NPU | Depth-Anything.tflite |
Depth-Anything | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 287.241 ms | 0 - 259 MB | FP16 | NPU | Depth-Anything.so |
Depth-Anything | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 196.175 ms | 2 - 990 MB | FP16 | NPU | Depth-Anything.onnx |
Depth-Anything | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 235.48 ms | 1 - 272 MB | FP16 | NPU | Depth-Anything.tflite |
Depth-Anything | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 274.953 ms | 3 - 283 MB | FP16 | NPU | Use Export Script |
Depth-Anything | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 155.528 ms | 0 - 520 MB | FP16 | NPU | Depth-Anything.onnx |
Depth-Anything | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 330.925 ms | 0 - 53 MB | FP16 | NPU | Depth-Anything.tflite |
Depth-Anything | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 240.606 ms | 4 - 5 MB | FP16 | NPU | Use Export Script |
Depth-Anything | SA7255P ADP | SA7255P | TFLITE | 1138.385 ms | 0 - 269 MB | FP16 | NPU | Depth-Anything.tflite |
Depth-Anything | SA7255P ADP | SA7255P | QNN | 1009.095 ms | 3 - 13 MB | FP16 | NPU | Use Export Script |
Depth-Anything | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 320.338 ms | 0 - 119 MB | FP16 | NPU | Depth-Anything.tflite |
Depth-Anything | SA8255 (Proxy) | SA8255P Proxy | QNN | 228.116 ms | 7 - 8 MB | FP16 | NPU | Use Export Script |
Depth-Anything | SA8295P ADP | SA8295P | TFLITE | 388.239 ms | 1 - 272 MB | FP16 | NPU | Depth-Anything.tflite |
Depth-Anything | SA8295P ADP | SA8295P | QNN | 280.573 ms | 3 - 8 MB | FP16 | NPU | Use Export Script |
Depth-Anything | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 328.502 ms | 1 - 83 MB | FP16 | NPU | Depth-Anything.tflite |
Depth-Anything | SA8650 (Proxy) | SA8650P Proxy | QNN | 228.865 ms | 4 - 5 MB | FP16 | NPU | Use Export Script |
Depth-Anything | SA8775P ADP | SA8775P | TFLITE | 368.799 ms | 1 - 270 MB | FP16 | NPU | Depth-Anything.tflite |
Depth-Anything | SA8775P ADP | SA8775P | QNN | 263.829 ms | 2 - 12 MB | FP16 | NPU | Use Export Script |
Depth-Anything | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 372.597 ms | 1 - 260 MB | FP16 | NPU | Depth-Anything.tflite |
Depth-Anything | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 430.273 ms | 0 - 276 MB | FP16 | NPU | Use Export Script |
Depth-Anything | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 212.934 ms | 3 - 3 MB | FP16 | NPU | Use Export Script |
Depth-Anything | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 272.145 ms | 64 - 64 MB | FP16 | NPU | Depth-Anything.onnx |
Installation
This model can be installed as a Python package via pip.
pip install "qai-hub-models[depth_anything]"
Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
Sign-in to Qualcomm® AI Hub with your
Qualcomm® ID. Once signed in navigate to Account -> Settings -> API Token
.
With this API token, you can configure your client to run models on the cloud hosted devices.
qai-hub configure --api_token API_TOKEN
Navigate to docs for more information.
Demo off target
The package contains a simple end-to-end demo that downloads pre-trained weights and runs this model on a sample input.
python -m qai_hub_models.models.depth_anything.demo
The above demo runs a reference implementation of pre-processing, model inference, and post processing.
NOTE: If you want running in a Jupyter Notebook or Google Colab like environment, please add the following to your cell (instead of the above).
%run -m qai_hub_models.models.depth_anything.demo
Run model on a cloud-hosted device
In addition to the demo, you can also run the model on a cloud-hosted Qualcomm® device. This script does the following:
- Performance check on-device on a cloud-hosted device
- Downloads compiled assets that can be deployed on-device for Android.
- Accuracy check between PyTorch and on-device outputs.
python -m qai_hub_models.models.depth_anything.export
Profiling Results
------------------------------------------------------------
Depth-Anything
Device : Samsung Galaxy S23 (13)
Runtime : TFLITE
Estimated inference time (ms) : 329.1
Estimated peak memory usage (MB): [0, 107]
Total # Ops : 635
Compute Unit(s) : NPU (635 ops)
How does this work?
This export script leverages Qualcomm® AI Hub to optimize, validate, and deploy this model on-device. Lets go through each step below in detail:
Step 1: Compile model for on-device deployment
To compile a PyTorch model for on-device deployment, we first trace the model
in memory using the jit.trace
and then call the submit_compile_job
API.
import torch
import qai_hub as hub
from qai_hub_models.models.depth_anything import Model
# Load the model
torch_model = Model.from_pretrained()
# Device
device = hub.Device("Samsung Galaxy S23")
# Trace model
input_shape = torch_model.get_input_spec()
sample_inputs = torch_model.sample_inputs()
pt_model = torch.jit.trace(torch_model, [torch.tensor(data[0]) for _, data in sample_inputs.items()])
# Compile model on a specific device
compile_job = hub.submit_compile_job(
model=pt_model,
device=device,
input_specs=torch_model.get_input_spec(),
)
# Get target model to run on-device
target_model = compile_job.get_target_model()
Step 2: Performance profiling on cloud-hosted device
After compiling models from step 1. Models can be profiled model on-device using the
target_model
. Note that this scripts runs the model on a device automatically
provisioned in the cloud. Once the job is submitted, you can navigate to a
provided job URL to view a variety of on-device performance metrics.
profile_job = hub.submit_profile_job(
model=target_model,
device=device,
)
Step 3: Verify on-device accuracy
To verify the accuracy of the model on-device, you can run on-device inference on sample input data on the same cloud hosted device.
input_data = torch_model.sample_inputs()
inference_job = hub.submit_inference_job(
model=target_model,
device=device,
inputs=input_data,
)
on_device_output = inference_job.download_output_data()
With the output of the model, you can compute like PSNR, relative errors or spot check the output with expected output.
Note: This on-device profiling and inference requires access to Qualcomm® AI Hub. Sign up for access.
Run demo on a cloud-hosted device
You can also run the demo on-device.
python -m qai_hub_models.models.depth_anything.demo --on-device
NOTE: If you want running in a Jupyter Notebook or Google Colab like environment, please add the following to your cell (instead of the above).
%run -m qai_hub_models.models.depth_anything.demo -- --on-device
Deploying compiled model to Android
The models can be deployed using multiple runtimes:
TensorFlow Lite (
.tflite
export): This tutorial provides a guide to deploy the .tflite model in an Android application.QNN (
.so
export ): This sample app provides instructions on how to use the.so
shared library in an Android application.
View on Qualcomm® AI Hub
Get more details on Depth-Anything's performance across various devices here. Explore all available models on Qualcomm® AI Hub
License
- The license for the original implementation of Depth-Anything can be found here.
- The license for the compiled assets for on-device deployment 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.