Text

Vision Transformer for Fast and Efficient Scene Text Recognition

deep-text-recognition-benchmark

ViTSTR is a simple single-stage model that uses a pre-trained Vision Transformer (ViT) to perform Scene Text Recognition (ViTSTR). It has a comparable accuracy with state-of-the-art STR models although it uses significantly less number of parameters and FLOPS. ViTSTR is also fast due to the parallel computation inherent to ViT architecture.

deep-text-recognition-benchmark

ViTSTR is built using a fork of CLOVA AI Deep Text Recognition Benchmark whose original documentation is at the bottom. Below we document how to train and evaluate ViTSTR-Tiny and ViTSTR-small.

Install requirements

pip3 install -r requirements.txt

Dataset

Download lmdb dataset. See CLOVA AI original documentation below.

Quick validation using a pre-trained model

ViTSTR-Small

CUDA_VISIBLE_DEVICES=0 python3 test.py --eval_data data_lmdb_release/evaluation

--benchmark_all_eval --Transformation None --FeatureExtraction None

--SequenceModeling None --Prediction None --Transformer

--sensitive


 


 




 


To finish reading, please visit source site