TSAL: Few-shot Text Segmentation Based on Attribute Learning
This addresses the high cost of pixel annotation in scene text segmentation for computer vision applications, though it is incremental as it builds on existing few-shot learning and CLIP methods.
The paper tackles the problem of scene text segmentation with limited labeled data by proposing TSAL, a few-shot learning method that leverages CLIP's prior knowledge and adaptive prompts to learn text attributes, achieving state-of-the-art performance on multiple datasets under few-shot settings.
Recently supervised learning rapidly develops in scene text segmentation. However, the lack of high-quality datasets and the high cost of pixel annotation greatly limit the development of them. Considering the well-performed few-shot learning methods for downstream tasks, we investigate the application of the few-shot learning method to scene text segmentation. We propose TSAL, which leverages CLIP's prior knowledge to learn text attributes for segmentation. To fully utilize the semantic and texture information in the image, a visual-guided branch is proposed to separately extract text and background features. To reduce data dependency and improve text detection accuracy, the adaptive prompt-guided branch employs effective adaptive prompt templates to capture various text attributes. To enable adaptive prompts capture distinctive text features and complex background distribution, we propose Adaptive Feature Alignment module(AFA). By aligning learnable tokens of different attributes with visual features and prompt prototypes, AFA enables adaptive prompts to capture both general and distinctive attribute information. TSAL can capture the unique attributes of text and achieve precise segmentation using only few images. Experiments demonstrate that our method achieves SOTA performance on multiple text segmentation datasets under few-shot settings and show great potential in text-related domains.