Hugging Face's logo

qualcomm
/
Depth-Anything

Depth Estimation
PyTorch
LiteRT
ONNX
android
Model card Files Community
Depth-Anything / README.md
qaihm-bot's picture
qaihm-bot
Upload README.md with huggingface_hub
f669895 verified
preview code
|
raw
history blame contribute delete
13.6 kB
metadata
library_name: pytorch
license: other
tags:
  - android
pipeline_tag: depth-estimation

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: Model_use_case.depth_estimation
  • Model Stats:
    • Model checkpoint: DepthAnything_Small
    • Input resolution: 518x518
    • Number of parameters: 24.8M
    • Model size: 94 MB
Model Precision Device Chipset Target Runtime Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit Target Model
Depth-Anything float QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) TFLITE 889.207 ms 1 - 585 MB NPU Depth-Anything.tflite
Depth-Anything float QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) QNN 272.538 ms 0 - 10 MB NPU Use Export Script
Depth-Anything float QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) TFLITE 194.929 ms 1 - 617 MB NPU Depth-Anything.tflite
Depth-Anything float QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) QNN 281.951 ms 2 - 663 MB NPU Use Export Script
Depth-Anything float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) TFLITE 136.87 ms 1 - 94 MB NPU Depth-Anything.tflite
Depth-Anything float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) QNN 140.076 ms 3 - 6 MB NPU Use Export Script
Depth-Anything float QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) TFLITE 141.898 ms 1 - 585 MB NPU Depth-Anything.tflite
Depth-Anything float QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) QNN 143.261 ms 0 - 11 MB NPU Use Export Script
Depth-Anything float SA7255P ADP Qualcomm® SA7255P TFLITE 889.207 ms 1 - 585 MB NPU Depth-Anything.tflite
Depth-Anything float SA7255P ADP Qualcomm® SA7255P QNN 272.538 ms 0 - 10 MB NPU Use Export Script
Depth-Anything float SA8255 (Proxy) Qualcomm® SA8255P (Proxy) TFLITE 143.601 ms 1 - 89 MB NPU Depth-Anything.tflite
Depth-Anything float SA8255 (Proxy) Qualcomm® SA8255P (Proxy) QNN 142.411 ms 3 - 5 MB NPU Use Export Script
Depth-Anything float SA8295P ADP Qualcomm® SA8295P TFLITE 192.592 ms 1 - 589 MB NPU Depth-Anything.tflite
Depth-Anything float SA8295P ADP Qualcomm® SA8295P QNN 177.255 ms 0 - 18 MB NPU Use Export Script
Depth-Anything float SA8650 (Proxy) Qualcomm® SA8650P (Proxy) TFLITE 146.041 ms 1 - 91 MB NPU Depth-Anything.tflite
Depth-Anything float SA8650 (Proxy) Qualcomm® SA8650P (Proxy) QNN 143.912 ms 3 - 5 MB NPU Use Export Script
Depth-Anything float SA8775P ADP Qualcomm® SA8775P TFLITE 141.898 ms 1 - 585 MB NPU Depth-Anything.tflite
Depth-Anything float SA8775P ADP Qualcomm® SA8775P QNN 143.261 ms 0 - 11 MB NPU Use Export Script
Depth-Anything float Samsung Galaxy S23 Snapdragon® 8 Gen 2 Mobile TFLITE 137.176 ms 1 - 91 MB NPU Depth-Anything.tflite
Depth-Anything float Samsung Galaxy S23 Snapdragon® 8 Gen 2 Mobile QNN 146.161 ms 3 - 93 MB NPU Use Export Script
Depth-Anything float Samsung Galaxy S23 Snapdragon® 8 Gen 2 Mobile ONNX 149.58 ms 0 - 202 MB NPU Depth-Anything.onnx
Depth-Anything float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile TFLITE 109.044 ms 0 - 574 MB NPU Depth-Anything.tflite
Depth-Anything float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile QNN 109.468 ms 3 - 656 MB NPU Use Export Script
Depth-Anything float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile ONNX 112.089 ms 0 - 741 MB NPU Depth-Anything.onnx
Depth-Anything float Snapdragon 8 Elite QRD Snapdragon® 8 Elite Mobile TFLITE 96.942 ms 1 - 551 MB NPU Depth-Anything.tflite
Depth-Anything float Snapdragon 8 Elite QRD Snapdragon® 8 Elite Mobile QNN 98.111 ms 0 - 670 MB NPU Use Export Script
Depth-Anything float Snapdragon 8 Elite QRD Snapdragon® 8 Elite Mobile ONNX 98.647 ms 5 - 749 MB NPU Depth-Anything.onnx
Depth-Anything float Snapdragon X Elite CRD Snapdragon® X Elite QNN 133.059 ms 3 - 3 MB NPU Use Export Script
Depth-Anything float Snapdragon X Elite CRD Snapdragon® X Elite ONNX 148.016 ms 66 - 66 MB NPU Depth-Anything.onnx
Depth-Anything w8a16 Samsung Galaxy S23 Snapdragon® 8 Gen 2 Mobile ONNX 428.557 ms 94 - 319 MB NPU Depth-Anything.onnx
Depth-Anything w8a16 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile ONNX 346.723 ms 163 - 1148 MB NPU Depth-Anything.onnx
Depth-Anything w8a16 Snapdragon 8 Elite QRD Snapdragon® 8 Elite Mobile ONNX 342.126 ms 160 - 1165 MB NPU Depth-Anything.onnx
Depth-Anything w8a16 Snapdragon X Elite CRD Snapdragon® X Elite ONNX 433.112 ms 232 - 232 MB NPU Depth-Anything.onnx

Installation

Install the 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                          : cs_8275 (ANDROID 14)                 
Runtime                         : TFLITE                               
Estimated inference time (ms)   : 889.2                                
Estimated peak memory usage (MB): [1, 585]                             
Total # Ops                     : 646                                  
Compute Unit(s)                 : npu (646 ops) gpu (0 ops) cpu (0 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 S24")

# 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