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: 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 | 272.53 ms | 1 - 38 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 66.81 ms | 1 - 50 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 80.243 ms | 6 - 57 MB | NPU | EasyOCR.dlc |
| EasyOCRDetector | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 38.37 ms | 0 - 146 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 39.314 ms | 6 - 22 MB | NPU | EasyOCR.dlc |
| EasyOCRDetector | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 37.452 ms | 7 - 18 MB | NPU | EasyOCR.onnx.zip |
| EasyOCRDetector | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 70.066 ms | 1 - 38 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 272.53 ms | 1 - 38 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 38.798 ms | 0 - 156 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 39.793 ms | 6 - 22 MB | NPU | EasyOCR.dlc |
| EasyOCRDetector | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 76.792 ms | 1 - 43 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 77.398 ms | 6 - 55 MB | NPU | EasyOCR.dlc |
| EasyOCRDetector | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 38.42 ms | 0 - 156 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 39.15 ms | 6 - 23 MB | NPU | EasyOCR.dlc |
| EasyOCRDetector | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 70.066 ms | 1 - 38 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 28.243 ms | 1 - 51 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 28.089 ms | 6 - 51 MB | NPU | EasyOCR.dlc |
| EasyOCRDetector | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 27.743 ms | 7 - 49 MB | NPU | EasyOCR.onnx.zip |
| EasyOCRDetector | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 22.965 ms | 0 - 42 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 22.793 ms | 6 - 46 MB | NPU | EasyOCR.dlc |
| EasyOCRDetector | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 22.175 ms | 2 - 41 MB | NPU | EasyOCR.onnx.zip |
| EasyOCRDetector | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | TFLITE | 16.975 ms | 0 - 43 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 19.389 ms | 6 - 46 MB | NPU | EasyOCR.dlc |
| EasyOCRDetector | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | ONNX | 16.63 ms | 7 - 50 MB | NPU | EasyOCR.onnx.zip |
| EasyOCRDetector | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 40.704 ms | 28 - 28 MB | NPU | EasyOCR.dlc |
| EasyOCRDetector | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 38.468 ms | 35 - 35 MB | NPU | EasyOCR.onnx.zip |
| EasyOCRRecognizer | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 551.897 ms | 11 - 21 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 122.133 ms | 9 - 30 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 38.122 ms | 0 - 199 MB | NPU | EasyOCR.dlc |
| EasyOCRRecognizer | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 104.879 ms | 7 - 10 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 24.677 ms | 0 - 131 MB | NPU | EasyOCR.dlc |
| EasyOCRRecognizer | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 35.394 ms | 0 - 68 MB | NPU | EasyOCR.onnx.zip |
| EasyOCRRecognizer | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 418.9 ms | 10 - 20 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 551.897 ms | 11 - 21 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 102.795 ms | 9 - 11 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 24.911 ms | 0 - 115 MB | NPU | EasyOCR.dlc |
| EasyOCRRecognizer | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 215.237 ms | 6 - 23 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 40.899 ms | 0 - 196 MB | NPU | EasyOCR.dlc |
| EasyOCRRecognizer | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 102.089 ms | 8 - 11 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 24.897 ms | 0 - 121 MB | NPU | EasyOCR.dlc |
| EasyOCRRecognizer | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 418.9 ms | 10 - 20 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 97.855 ms | 9 - 29 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 18.988 ms | 0 - 465 MB | NPU | EasyOCR.dlc |
| EasyOCRRecognizer | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 26.376 ms | 0 - 270 MB | NPU | EasyOCR.onnx.zip |
| EasyOCRRecognizer | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 120.852 ms | 9 - 24 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 16.403 ms | 0 - 474 MB | NPU | EasyOCR.dlc |
| EasyOCRRecognizer | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 21.522 ms | 0 - 278 MB | NPU | EasyOCR.onnx.zip |
| EasyOCRRecognizer | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | TFLITE | 72.98 ms | 14 - 25 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 17.374 ms | 0 - 510 MB | NPU | EasyOCR.dlc |
| EasyOCRRecognizer | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | ONNX | 19.43 ms | 0 - 304 MB | NPU | EasyOCR.onnx.zip |
| EasyOCRRecognizer | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 24.838 ms | 82 - 82 MB | NPU | EasyOCR.dlc |
| EasyOCRRecognizer | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 35.981 ms | 11 - 11 MB | NPU | EasyOCR.onnx.zip |
| EasyOCRDetector | w8a8 | Dragonwing RB3 Gen 2 Vision Kit | Qualcomm® QCS6490 | TFLITE | 47.636 ms | 0 - 24 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 38.013 ms | 0 - 31 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 13.572 ms | 0 - 62 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 10.911 ms | 0 - 80 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 11.465 ms | 0 - 31 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | w8a8 | RB5 (Proxy) | Qualcomm® QCS8250 (Proxy) | TFLITE | 458.672 ms | 32 - 35 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | w8a8 | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 38.013 ms | 0 - 31 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | w8a8 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 10.953 ms | 0 - 78 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | w8a8 | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 20.485 ms | 0 - 36 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | w8a8 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 10.921 ms | 0 - 71 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | w8a8 | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 11.465 ms | 0 - 31 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 7.995 ms | 0 - 61 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 6.48 ms | 0 - 38 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | w8a8 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | TFLITE | 20.363 ms | 0 - 54 MB | NPU | EasyOCR.tflite |
| EasyOCRDetector | w8a8 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | TFLITE | 5.099 ms | 0 - 34 MB | NPU | EasyOCR.tflite |
| EasyOCRRecognizer | w8a8 | Dragonwing RB3 Gen 2 Vision Kit | Qualcomm® QCS6490 | TFLITE | 202.156 ms | 4 - 25 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 494.52 ms | 9 - 19 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 87.258 ms | 7 - 30 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 74.952 ms | 7 - 10 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 325.497 ms | 10 - 20 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | w8a8 | RB5 (Proxy) | Qualcomm® QCS8250 (Proxy) | TFLITE | 182.941 ms | 6 - 8 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | w8a8 | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 494.52 ms | 9 - 19 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | w8a8 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 77.825 ms | 5 - 15 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | w8a8 | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 193.683 ms | 8 - 25 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | w8a8 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 81.23 ms | 6 - 10 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | w8a8 | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 325.497 ms | 10 - 20 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 71.616 ms | 7 - 26 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 95.022 ms | 10 - 26 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | w8a8 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | TFLITE | 125.632 ms | 13 - 29 MB | CPU | EasyOCR.tflite |
| EasyOCRRecognizer | w8a8 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | TFLITE | 60.273 ms | 14 - 26 MB | CPU | EasyOCR.tflite |
Installation
Install the package via pip:
# NOTE: 3.10 <= PYTHON_VERSION < 3.14 is supported.
pip install "qai-hub-models[easyocr]"
Configure Qualcomm® AI Hub Workbench to run this model on a cloud-hosted device
Sign-in to Qualcomm® AI Hub Workbench 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 Workbench. 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.