Hugging Face's logo

qualcomm
/
TrOCR

Image-to-Text
PyTorch
android
Model card Files
xet
 Community
3
TrOCR / README.md
qaihm-bot's picture
qaihm-bot
Upload README.md with huggingface_hub
5291c19 verified
preview code
|
raw
history blame
15.4 kB
metadata 
library_name: pytorch
license: mit
pipeline_tag: image-to-text
tags:
  - android

TrOCR: Optimized for Mobile Deployment

Transformer based model for state-of-the-art optical character recognition (OCR) on both printed and handwritten text

End-to-end text recognition approach with pre-trained image transformer and text transformer models for both image understanding and wordpiece-level text generation.

This model is an implementation of TrOCR found here. This repository provides scripts to run TrOCR on Qualcomm® devices. More details on model performance across various devices, can be found here.

Model Details

  • Model Type: Image to text
  • Model Stats:
    • Model checkpoint: trocr-small-stage1
    • Input resolution: 320x320
    • Number of parameters (TrOCREncoder): 23.0M
    • Model size (TrOCREncoder): 87.8 MB
    • Number of parameters (TrOCRDecoder): 38.3M
    • Model size (TrOCRDecoder): 146 MB
Model Device Chipset Target Runtime Inference Time (ms) Peak Memory Range (MB) Precision Primary Compute Unit Target Model
TrOCREncoder Samsung Galaxy S23 Snapdragon® 8 Gen 2 TFLITE 50.652 ms 7 - 9 MB FP16 NPU TrOCR.tflite
TrOCREncoder Samsung Galaxy S23 Snapdragon® 8 Gen 2 QNN 52.89 ms 0 - 22 MB FP16 NPU TrOCR.so
TrOCREncoder Samsung Galaxy S23 Snapdragon® 8 Gen 2 ONNX 39.309 ms 0 - 178 MB FP16 NPU TrOCR.onnx
TrOCREncoder Samsung Galaxy S24 Snapdragon® 8 Gen 3 TFLITE 40.349 ms 5 - 306 MB FP16 NPU TrOCR.tflite
TrOCREncoder Samsung Galaxy S24 Snapdragon® 8 Gen 3 QNN 42.073 ms 2 - 64 MB FP16 NPU TrOCR.so
TrOCREncoder Samsung Galaxy S24 Snapdragon® 8 Gen 3 ONNX 31.228 ms 0 - 348 MB FP16 NPU TrOCR.onnx
TrOCREncoder QCS8550 (Proxy) QCS8550 Proxy TFLITE 50.061 ms 7 - 8 MB FP16 NPU TrOCR.tflite
TrOCREncoder QCS8550 (Proxy) QCS8550 Proxy QNN 36.086 ms 2 - 3 MB FP16 NPU Use Export Script
TrOCREncoder SA8255 (Proxy) SA8255P Proxy TFLITE 50.179 ms 7 - 9 MB FP16 NPU TrOCR.tflite
TrOCREncoder SA8255 (Proxy) SA8255P Proxy QNN 36.899 ms 2 - 4 MB FP16 NPU Use Export Script
TrOCREncoder SA8775 (Proxy) SA8775P Proxy TFLITE 51.951 ms 7 - 9 MB FP16 NPU TrOCR.tflite
TrOCREncoder SA8775 (Proxy) SA8775P Proxy QNN 37.124 ms 2 - 3 MB FP16 NPU Use Export Script
TrOCREncoder SA8650 (Proxy) SA8650P Proxy TFLITE 51.056 ms 7 - 9 MB FP16 NPU TrOCR.tflite
TrOCREncoder SA8650 (Proxy) SA8650P Proxy QNN 37.072 ms 2 - 7 MB FP16 NPU Use Export Script
TrOCREncoder QCS8450 (Proxy) QCS8450 Proxy TFLITE 60.938 ms 7 - 296 MB FP16 NPU TrOCR.tflite
TrOCREncoder QCS8450 (Proxy) QCS8450 Proxy QNN 60.192 ms 2 - 64 MB FP16 NPU Use Export Script
TrOCREncoder Snapdragon 8 Elite QRD Snapdragon® 8 Elite TFLITE 36.174 ms 3 - 119 MB FP16 NPU TrOCR.tflite
TrOCREncoder Snapdragon 8 Elite QRD Snapdragon® 8 Elite QNN 33.016 ms 2 - 66 MB FP16 NPU Use Export Script
TrOCREncoder Snapdragon 8 Elite QRD Snapdragon® 8 Elite ONNX 23.693 ms 6 - 207 MB FP16 NPU TrOCR.onnx
TrOCREncoder Snapdragon X Elite CRD Snapdragon® X Elite QNN 33.885 ms 2 - 2 MB FP16 NPU Use Export Script
TrOCREncoder Snapdragon X Elite CRD Snapdragon® X Elite ONNX 35.659 ms 109 - 109 MB FP16 NPU TrOCR.onnx
TrOCRDecoder Samsung Galaxy S23 Snapdragon® 8 Gen 2 TFLITE 2.6 ms 0 - 2 MB FP16 NPU TrOCR.tflite
TrOCRDecoder Samsung Galaxy S23 Snapdragon® 8 Gen 2 QNN 3.012 ms 3 - 263 MB FP16 NPU TrOCR.so
TrOCRDecoder Samsung Galaxy S23 Snapdragon® 8 Gen 2 ONNX 2.843 ms 1 - 3 MB FP16 NPU TrOCR.onnx
TrOCRDecoder Samsung Galaxy S24 Snapdragon® 8 Gen 3 TFLITE 1.851 ms 0 - 190 MB FP16 NPU TrOCR.tflite
TrOCRDecoder Samsung Galaxy S24 Snapdragon® 8 Gen 3 QNN 2.471 ms 0 - 51 MB FP16 NPU TrOCR.so
TrOCRDecoder Samsung Galaxy S24 Snapdragon® 8 Gen 3 ONNX 2.148 ms 0 - 149 MB FP16 NPU TrOCR.onnx
TrOCRDecoder QCS8550 (Proxy) QCS8550 Proxy TFLITE 2.562 ms 0 - 2 MB FP16 NPU TrOCR.tflite
TrOCRDecoder QCS8550 (Proxy) QCS8550 Proxy QNN 2.631 ms 0 - 1 MB FP16 NPU Use Export Script
TrOCRDecoder SA8255 (Proxy) SA8255P Proxy TFLITE 2.608 ms 0 - 2 MB FP16 NPU TrOCR.tflite
TrOCRDecoder SA8255 (Proxy) SA8255P Proxy QNN 2.607 ms 1 - 3 MB FP16 NPU Use Export Script
TrOCRDecoder SA8775 (Proxy) SA8775P Proxy TFLITE 2.604 ms 0 - 2 MB FP16 NPU TrOCR.tflite
TrOCRDecoder SA8775 (Proxy) SA8775P Proxy QNN 2.613 ms 1 - 3 MB FP16 NPU Use Export Script
TrOCRDecoder SA8650 (Proxy) SA8650P Proxy TFLITE 2.573 ms 0 - 2 MB FP16 NPU TrOCR.tflite
TrOCRDecoder SA8650 (Proxy) SA8650P Proxy QNN 2.658 ms 1 - 3 MB FP16 NPU Use Export Script
TrOCRDecoder QCS8450 (Proxy) QCS8450 Proxy TFLITE 2.814 ms 0 - 189 MB FP16 NPU TrOCR.tflite
TrOCRDecoder QCS8450 (Proxy) QCS8450 Proxy QNN 3.375 ms 4 - 52 MB FP16 NPU Use Export Script
TrOCRDecoder Snapdragon 8 Elite QRD Snapdragon® 8 Elite TFLITE 2.104 ms 0 - 27 MB FP16 NPU TrOCR.tflite
TrOCRDecoder Snapdragon 8 Elite QRD Snapdragon® 8 Elite QNN 2.016 ms 0 - 45 MB FP16 NPU Use Export Script
TrOCRDecoder Snapdragon 8 Elite QRD Snapdragon® 8 Elite ONNX 2.078 ms 0 - 35 MB FP16 NPU TrOCR.onnx
TrOCRDecoder Snapdragon X Elite CRD Snapdragon® X Elite QNN 2.793 ms 7 - 7 MB FP16 NPU Use Export Script
TrOCRDecoder Snapdragon X Elite CRD Snapdragon® X Elite ONNX 2.881 ms 68 - 68 MB FP16 NPU TrOCR.onnx

Installation

This model can be installed as a Python package via pip.

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

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.trocr.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.trocr.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.trocr.export

Profiling Results
------------------------------------------------------------
TrOCREncoder
Device                          : Samsung Galaxy S23 (13)
Runtime                         : TFLITE                 
Estimated inference time (ms)   : 50.7                   
Estimated peak memory usage (MB): [7, 9]                 
Total # Ops                     : 591                    
Compute Unit(s)                 : NPU (591 ops)          

------------------------------------------------------------
TrOCRDecoder
Device                          : Samsung Galaxy S23 (13)
Runtime                         : TFLITE                 
Estimated inference time (ms)   : 2.6                    
Estimated peak memory usage (MB): [0, 2]                 
Total # Ops                     : 399                    
Compute Unit(s)                 : NPU (399 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.trocr import TrOCREncoder,TrOCRDecoder

# Load the model
encoder_model = TrOCREncoder.from_pretrained()
decoder_model = TrOCRDecoder.from_pretrained()

# Device
device = hub.Device("Samsung Galaxy S23")

# Trace model
encoder_input_shape = encoder_model.get_input_spec()
encoder_sample_inputs = encoder_model.sample_inputs()

traced_encoder_model = torch.jit.trace(encoder_model, [torch.tensor(data[0]) for _, data in encoder_sample_inputs.items()])

# Compile model on a specific device
encoder_compile_job = hub.submit_compile_job(
    model=traced_encoder_model ,
    device=device,
    input_specs=encoder_model.get_input_spec(),
)

# Get target model to run on-device
encoder_target_model = encoder_compile_job.get_target_model()
# Trace model
decoder_input_shape = decoder_model.get_input_spec()
decoder_sample_inputs = decoder_model.sample_inputs()

traced_decoder_model = torch.jit.trace(decoder_model, [torch.tensor(data[0]) for _, data in decoder_sample_inputs.items()])

# Compile model on a specific device
decoder_compile_job = hub.submit_compile_job(
    model=traced_decoder_model ,
    device=device,
    input_specs=decoder_model.get_input_spec(),
)

# Get target model to run on-device
decoder_target_model = decoder_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.

encoder_profile_job = hub.submit_profile_job(
    model=encoder_target_model,
    device=device,
)
decoder_profile_job = hub.submit_profile_job(
    model=decoder_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.

encoder_input_data = encoder_model.sample_inputs()
encoder_inference_job = hub.submit_inference_job(
    model=encoder_target_model,
    device=device,
    inputs=encoder_input_data,
)
encoder_inference_job.download_output_data()
decoder_input_data = decoder_model.sample_inputs()
decoder_inference_job = hub.submit_inference_job(
    model=decoder_target_model,
    device=device,
    inputs=decoder_input_data,
)
decoder_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 TrOCR's performance across various devices here. Explore all available models on Qualcomm® AI Hub

License

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

References

Community