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: 608x800
- Number of parameters (EasyOCRDetector): 20.8M
- Model size (EasyOCRDetector) (float): 79.2 MB
- Number of parameters (EasyOCRRecognizer): 3.84M
- Model size (EasyOCRRecognizer) (float): 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 | 272.521 ms | 1 - 38 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 66.695 ms | 1 - 47 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 80.417 ms | 6 - 56 MB | NPU | EasyOCR.dlc |
| EasyOCRDetector | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 38.318 ms | 0 - 154 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 39.369 ms | 6 - 23 MB | NPU | EasyOCR.dlc |
| EasyOCRDetector | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 37.689 ms | 0 - 52 MB | NPU | EasyOCR.onnx.zip |
| EasyOCRDetector | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 70.007 ms | 13 - 49 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 272.521 ms | 1 - 38 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 38.198 ms | 1 - 154 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 39.829 ms | 6 - 20 MB | NPU | EasyOCR.dlc |
| EasyOCRDetector | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 76.771 ms | 1 - 40 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 77.272 ms | 6 - 53 MB | NPU | EasyOCR.dlc |
| EasyOCRDetector | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 39.149 ms | 0 - 145 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 39.436 ms | 6 - 22 MB | NPU | EasyOCR.dlc |
| EasyOCRDetector | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 70.007 ms | 13 - 49 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 28.195 ms | 1 - 45 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 28.584 ms | 6 - 46 MB | NPU | EasyOCR.dlc |
| EasyOCRDetector | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 27.276 ms | 6 - 48 MB | NPU | EasyOCR.onnx.zip |
| EasyOCRDetector | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 22.79 ms | 1 - 42 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 22.779 ms | 6 - 44 MB | NPU | EasyOCR.dlc |
| EasyOCRDetector | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 22.343 ms | 4 - 44 MB | NPU | EasyOCR.onnx.zip |
| EasyOCRDetector | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | TFLITE | 17.02 ms | 1 - 42 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | ONNX | 17.664 ms | 7 - 49 MB | NPU | EasyOCR.onnx.zip |
| EasyOCRDetector | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 40.778 ms | 34 - 34 MB | NPU | EasyOCR.dlc |
| EasyOCRDetector | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 38.495 ms | 35 - 35 MB | NPU | EasyOCR.onnx.zip |
| EasyOCRRecognizer | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 479.613 ms | 8 - 18 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 113.782 ms | 9 - 30 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 38.101 ms | 0 - 196 MB | NPU | EasyOCR.dlc |
| EasyOCRRecognizer | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 104.991 ms | 8 - 11 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 24.676 ms | 0 - 103 MB | NPU | EasyOCR.dlc |
| EasyOCRRecognizer | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 35.392 ms | 0 - 86 MB | NPU | EasyOCR.onnx.zip |
| EasyOCRRecognizer | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 352.662 ms | 8 - 18 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 479.613 ms | 8 - 18 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 109.667 ms | 7 - 10 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 24.736 ms | 0 - 102 MB | NPU | EasyOCR.dlc |
| EasyOCRRecognizer | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 208.124 ms | 8 - 26 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 40.581 ms | 0 - 194 MB | NPU | EasyOCR.dlc |
| EasyOCRRecognizer | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 102.007 ms | 6 - 9 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 24.801 ms | 0 - 103 MB | NPU | EasyOCR.dlc |
| EasyOCRRecognizer | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 352.662 ms | 8 - 18 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 93.664 ms | 9 - 29 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 18.627 ms | 0 - 464 MB | NPU | EasyOCR.dlc |
| EasyOCRRecognizer | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 26.499 ms | 0 - 273 MB | NPU | EasyOCR.onnx.zip |
| EasyOCRRecognizer | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 111.258 ms | 11 - 23 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 16.059 ms | 0 - 473 MB | NPU | EasyOCR.dlc |
| EasyOCRRecognizer | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 21.397 ms | 0 - 279 MB | NPU | EasyOCR.onnx.zip |
| EasyOCRRecognizer | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | TFLITE | 84.689 ms | 13 - 25 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | ONNX | 19.914 ms | 0 - 302 MB | NPU | EasyOCR.onnx.zip |
| EasyOCRRecognizer | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 24.52 ms | 73 - 73 MB | NPU | EasyOCR.dlc |
| EasyOCRRecognizer | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 35.965 ms | 12 - 12 MB | NPU | EasyOCR.onnx.zip |
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
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 S25")
# 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 (
.tfliteexport): This tutorial provides a guide to deploy the .tflite model in an Android application.QNN (
.soexport ): This sample app provides instructions on how to use the.soshared 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
- 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.