Hugging Face's logo

qualcomm
/
EasyOCR

Image-to-Text
PyTorch
LiteRT
ONNX
android
Model card Files Community
EasyOCR / README.md
qaihm-bot's picture
qaihm-bot
v0.30.2
9d1adf6 verified
preview code
|
raw
history blame contribute delete
17.3 kB
metadata
library_name: pytorch
license: other
tags:
  - android
pipeline_tag: image-to-text

EasyOCR: Optimized for Mobile Deployment

Ready-to-use OCR with 80+ supported languages and all popular writing scripts

EasyOCR is a machine learning model that can recognize text in images. It supports 80+ supported languages and all popular writing scripts.

This model is an implementation of EasyOCR found here.

This repository provides scripts to run EasyOCR on Qualcomm® devices. More details on model performance across various devices, can be found here.

Model Details

  • Model Type: Model_use_case.image_to_text
  • Model Stats:
    • Model checkpoint: easyocr-small-stage1
    • Input resolution: 384x384
    • Number of parameters (EasyOCRDetector): 20.8M
    • Model size (EasyOCRDetector): 79.2 MB
    • Number of parameters (EasyOCRRecognizer): 3.84M
    • Model size (EasyOCRRecognizer): 14.7 MB
Model Precision Device Chipset Target Runtime Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit Target Model
EasyOCRDetector float QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) TFLITE 275.464 ms 16 - 45 MB NPU EasyOCR.tflite
EasyOCRDetector float QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) QNN 270.706 ms 1 - 11 MB NPU Use Export Script
EasyOCRDetector float QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) TFLITE 71.067 ms 16 - 76 MB NPU EasyOCR.tflite
EasyOCRDetector float QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) QNN 75.3 ms 6 - 44 MB NPU Use Export Script
EasyOCRDetector float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) TFLITE 40.653 ms 12 - 143 MB NPU EasyOCR.tflite
EasyOCRDetector float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) QNN 38.186 ms 6 - 8 MB NPU Use Export Script
EasyOCRDetector float QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) TFLITE 71.919 ms 16 - 46 MB NPU EasyOCR.tflite
EasyOCRDetector float QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) QNN 68.88 ms 2 - 16 MB NPU Use Export Script
EasyOCRDetector float SA7255P ADP Qualcomm® SA7255P TFLITE 275.464 ms 16 - 45 MB NPU EasyOCR.tflite
EasyOCRDetector float SA7255P ADP Qualcomm® SA7255P QNN 270.706 ms 1 - 11 MB NPU Use Export Script
EasyOCRDetector float SA8255 (Proxy) Qualcomm® SA8255P (Proxy) TFLITE 41.834 ms 10 - 142 MB NPU EasyOCR.tflite
EasyOCRDetector float SA8255 (Proxy) Qualcomm® SA8255P (Proxy) QNN 38.861 ms 6 - 8 MB NPU Use Export Script
EasyOCRDetector float SA8295P ADP Qualcomm® SA8295P TFLITE 78.541 ms 16 - 46 MB NPU EasyOCR.tflite
EasyOCRDetector float SA8295P ADP Qualcomm® SA8295P QNN 74.988 ms 0 - 18 MB NPU Use Export Script
EasyOCRDetector float SA8650 (Proxy) Qualcomm® SA8650P (Proxy) TFLITE 40.537 ms 11 - 143 MB NPU EasyOCR.tflite
EasyOCRDetector float SA8650 (Proxy) Qualcomm® SA8650P (Proxy) QNN 38.48 ms 6 - 8 MB NPU Use Export Script
EasyOCRDetector float SA8775P ADP Qualcomm® SA8775P TFLITE 71.919 ms 16 - 46 MB NPU EasyOCR.tflite
EasyOCRDetector float SA8775P ADP Qualcomm® SA8775P QNN 68.88 ms 2 - 16 MB NPU Use Export Script
EasyOCRDetector float Samsung Galaxy S23 Snapdragon® 8 Gen 2 Mobile TFLITE 41.171 ms 12 - 146 MB NPU EasyOCR.tflite
EasyOCRDetector float Samsung Galaxy S23 Snapdragon® 8 Gen 2 Mobile QNN 38.256 ms 2 - 17 MB NPU Use Export Script
EasyOCRDetector float Samsung Galaxy S23 Snapdragon® 8 Gen 2 Mobile ONNX 40.044 ms 33 - 123 MB NPU EasyOCR.onnx
EasyOCRDetector float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile TFLITE 30.148 ms 16 - 74 MB NPU EasyOCR.tflite
EasyOCRDetector float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile QNN 28.185 ms 6 - 41 MB NPU Use Export Script
EasyOCRDetector float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile ONNX 28.315 ms 34 - 69 MB NPU EasyOCR.onnx
EasyOCRDetector float Snapdragon 8 Elite QRD Snapdragon® 8 Elite Mobile TFLITE 24.527 ms 15 - 49 MB NPU EasyOCR.tflite
EasyOCRDetector float Snapdragon 8 Elite QRD Snapdragon® 8 Elite Mobile QNN 23.248 ms 6 - 40 MB NPU Use Export Script
EasyOCRDetector float Snapdragon 8 Elite QRD Snapdragon® 8 Elite Mobile ONNX 23.523 ms 38 - 71 MB NPU EasyOCR.onnx
EasyOCRDetector float Snapdragon X Elite CRD Snapdragon® X Elite QNN 38.631 ms 6 - 6 MB NPU Use Export Script
EasyOCRDetector float Snapdragon X Elite CRD Snapdragon® X Elite ONNX 40.317 ms 72 - 72 MB NPU EasyOCR.onnx
EasyOCRRecognizer float QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) TFLITE 483.485 ms 8 - 18 MB CPU EasyOCR.tflite
EasyOCRRecognizer float QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) QNN 61.815 ms 0 - 9 MB NPU Use Export Script
EasyOCRRecognizer float QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) TFLITE 298.408 ms 4 - 28 MB CPU EasyOCR.tflite
EasyOCRRecognizer float QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) QNN 37.653 ms 0 - 167 MB NPU Use Export Script
EasyOCRRecognizer float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) TFLITE 122.451 ms 4 - 8 MB CPU EasyOCR.tflite
EasyOCRRecognizer float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) QNN 23.22 ms 0 - 4 MB NPU Use Export Script
EasyOCRRecognizer float QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) TFLITE 354.941 ms 9 - 24 MB CPU EasyOCR.tflite
EasyOCRRecognizer float QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) QNN 27.702 ms 0 - 14 MB NPU Use Export Script
EasyOCRRecognizer float SA7255P ADP Qualcomm® SA7255P TFLITE 483.485 ms 8 - 18 MB CPU EasyOCR.tflite
EasyOCRRecognizer float SA7255P ADP Qualcomm® SA7255P QNN 61.815 ms 0 - 9 MB NPU Use Export Script
EasyOCRRecognizer float SA8255 (Proxy) Qualcomm® SA8255P (Proxy) TFLITE 111.415 ms 6 - 9 MB CPU EasyOCR.tflite
EasyOCRRecognizer float SA8255 (Proxy) Qualcomm® SA8255P (Proxy) QNN 23.194 ms 0 - 2 MB NPU Use Export Script
EasyOCRRecognizer float SA8295P ADP Qualcomm® SA8295P TFLITE 222.734 ms 10 - 28 MB CPU EasyOCR.tflite
EasyOCRRecognizer float SA8295P ADP Qualcomm® SA8295P QNN 39.068 ms 0 - 18 MB NPU Use Export Script
EasyOCRRecognizer float SA8650 (Proxy) Qualcomm® SA8650P (Proxy) TFLITE 127.195 ms 8 - 11 MB CPU EasyOCR.tflite
EasyOCRRecognizer float SA8650 (Proxy) Qualcomm® SA8650P (Proxy) QNN 23.286 ms 0 - 2 MB NPU Use Export Script
EasyOCRRecognizer float SA8775P ADP Qualcomm® SA8775P TFLITE 354.941 ms 9 - 24 MB CPU EasyOCR.tflite
EasyOCRRecognizer float SA8775P ADP Qualcomm® SA8775P QNN 27.702 ms 0 - 14 MB NPU Use Export Script
EasyOCRRecognizer float Samsung Galaxy S23 Snapdragon® 8 Gen 2 Mobile TFLITE 112.343 ms 3 - 5 MB CPU EasyOCR.tflite
EasyOCRRecognizer float Samsung Galaxy S23 Snapdragon® 8 Gen 2 Mobile QNN 24.296 ms 0 - 102 MB NPU Use Export Script
EasyOCRRecognizer float Samsung Galaxy S23 Snapdragon® 8 Gen 2 Mobile ONNX 22.138 ms 0 - 26 MB NPU EasyOCR.onnx
EasyOCRRecognizer float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile TFLITE 94.787 ms 9 - 29 MB CPU EasyOCR.tflite
EasyOCRRecognizer float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile QNN 17.766 ms 0 - 432 MB NPU Use Export Script
EasyOCRRecognizer float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile ONNX 15.721 ms 1 - 19 MB NPU EasyOCR.onnx
EasyOCRRecognizer float Snapdragon 8 Elite QRD Snapdragon® 8 Elite Mobile TFLITE 107.21 ms 20 - 34 MB CPU EasyOCR.tflite
EasyOCRRecognizer float Snapdragon 8 Elite QRD Snapdragon® 8 Elite Mobile QNN 18.672 ms 0 - 432 MB NPU Use Export Script
EasyOCRRecognizer float Snapdragon 8 Elite QRD Snapdragon® 8 Elite Mobile ONNX 16.837 ms 0 - 18 MB NPU EasyOCR.onnx
EasyOCRRecognizer float Snapdragon X Elite CRD Snapdragon® X Elite QNN 24.475 ms 0 - 0 MB NPU Use Export Script
EasyOCRRecognizer float Snapdragon X Elite CRD Snapdragon® X Elite ONNX 17.322 ms 0 - 0 MB NPU EasyOCR.onnx

Installation

Install the package via pip:

pip install "qai-hub-models[easyocr]"

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.easyocr.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.easyocr.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.easyocr.export

Profiling Results
------------------------------------------------------------
EasyOCRDetector
Device                          : cs_8275 (ANDROID 14)                
Runtime                         : TFLITE                              
Estimated inference time (ms)   : 275.5                               
Estimated peak memory usage (MB): [16, 45]                            
Total # Ops                     : 42                                  
Compute Unit(s)                 : npu (42 ops) gpu (0 ops) cpu (0 ops)

------------------------------------------------------------
EasyOCRRecognizer
Device                          : cs_8275 (ANDROID 14)                 
Runtime                         : TFLITE                               
Estimated inference time (ms)   : 483.5                                
Estimated peak memory usage (MB): [8, 18]                              
Total # Ops                     : 136                                  
Compute Unit(s)                 : npu (0 ops) gpu (0 ops) cpu (136 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.easyocr 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.

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 EasyOCR's performance across various devices here. Explore all available models on Qualcomm® AI Hub

License

  • The license for the original implementation of EasyOCR can be found here.
  • The license for the compiled assets for on-device deployment can be found here

References

Community