AutoSTR: Efficient Backbone Search for Scene Text Recognition
arXiv:2003.06567
Abstract
Scene text recognition (STR) is very challenging due to the diversity of text instances and the complexity of scenes. The community has paid increasing attention to boost the performance by improving the pre-processing image module, like rectification and deblurring, or the sequence translator. However, another critical module, i.e., the feature sequence extractor, has not been extensively explored. In this work, inspired by the success of neural architecture search (NAS), which can identify better architectures than human-designed ones, we propose automated STR (AutoSTR) to search data-dependent backbones to boost text recognition performance. First, we design a domain-specific search space for STR, which contains both choices on operations and constraints on the downsampling path. Then, we propose a two-step search algorithm, which decouples operations and downsampling path, for an efficient search in the given space. Experiments demonstrate that, by searching data-dependent backbones, AutoSTR can outperform the state-of-the-art approaches on standard benchmarks with much fewer FLOPS and model parameters.
arXiv commentECCV 2020
References in corpus (12)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Neural Machine Translation by Jointly Learning to Align and Translate
- ADADELTA: An Adaptive Learning Rate Method
- Neural Architecture Search with Reinforcement Learning
- Neural Architecture Search: A Survey
- DARTS: Differentiable Architecture Search
- Synthetic Data and Artificial Neural Networks for Natural Scene Text Recognition
- Efficient Neural Architecture Search via Parameter Sharing
- Focusing Attention: Towards Accurate Text Recognition in Natural Images
- ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware
- Random Search and Reproducibility for Neural Architecture Search
- DetNAS: Backbone Search for Object Detection
Cited by in corpus (4)
- TRIG: Transformer-Based Text Recognizer with Initial Embedding Guidance
- FAST: Faster Arbitrarily-Shaped Text Detector with Minimalist Kernel Representation
- Multi-modal Text Recognition Networks: Interactive Enhancements between Visual and Semantic Features
- Reciprocal Feature Learning via Explicit and Implicit Tasks in Scene Text Recognition