Char-SAM: Turning Segment Anything Model into Scene Text Segmentation Annotator with Character-level Visual Prompts

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xie, Enze, Lyu, Jiaho, Wu, Daiqing, Shen, Huawen, Zhou, Yu
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910766126333952
author Xie, Enze
Lyu, Jiaho
Wu, Daiqing
Shen, Huawen
Zhou, Yu
author_facet Xie, Enze
Lyu, Jiaho
Wu, Daiqing
Shen, Huawen
Zhou, Yu
contents The recent emergence of the Segment Anything Model (SAM) enables various domain-specific segmentation tasks to be tackled cost-effectively by using bounding boxes as prompts. However, in scene text segmentation, SAM can not achieve desirable performance. The word-level bounding box as prompts is too coarse for characters, while the character-level bounding box as prompts suffers from over-segmentation and under-segmentation issues. In this paper, we propose an automatic annotation pipeline named Char-SAM, that turns SAM into a low-cost segmentation annotator with a Character-level visual prompt. Specifically, leveraging some existing text detection datasets with word-level bounding box annotations, we first generate finer-grained character-level bounding box prompts using the Character Bounding-box Refinement CBR module. Next, we employ glyph information corresponding to text character categories as a new prompt in the Character Glyph Refinement (CGR) module to guide SAM in producing more accurate segmentation masks, addressing issues of over-segmentation and under-segmentation. These modules fully utilize the bbox-to-mask capability of SAM to generate high-quality text segmentation annotations automatically. Extensive experiments on TextSeg validate the effectiveness of Char-SAM. Its training-free nature also enables the generation of high-quality scene text segmentation datasets from real-world datasets like COCO-Text and MLT17.
format Preprint
id arxiv_https___arxiv_org_abs_2412_19917
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Char-SAM: Turning Segment Anything Model into Scene Text Segmentation Annotator with Character-level Visual Prompts
Xie, Enze
Lyu, Jiaho
Wu, Daiqing
Shen, Huawen
Zhou, Yu
Computer Vision and Pattern Recognition
The recent emergence of the Segment Anything Model (SAM) enables various domain-specific segmentation tasks to be tackled cost-effectively by using bounding boxes as prompts. However, in scene text segmentation, SAM can not achieve desirable performance. The word-level bounding box as prompts is too coarse for characters, while the character-level bounding box as prompts suffers from over-segmentation and under-segmentation issues. In this paper, we propose an automatic annotation pipeline named Char-SAM, that turns SAM into a low-cost segmentation annotator with a Character-level visual prompt. Specifically, leveraging some existing text detection datasets with word-level bounding box annotations, we first generate finer-grained character-level bounding box prompts using the Character Bounding-box Refinement CBR module. Next, we employ glyph information corresponding to text character categories as a new prompt in the Character Glyph Refinement (CGR) module to guide SAM in producing more accurate segmentation masks, addressing issues of over-segmentation and under-segmentation. These modules fully utilize the bbox-to-mask capability of SAM to generate high-quality text segmentation annotations automatically. Extensive experiments on TextSeg validate the effectiveness of Char-SAM. Its training-free nature also enables the generation of high-quality scene text segmentation datasets from real-world datasets like COCO-Text and MLT17.
title Char-SAM: Turning Segment Anything Model into Scene Text Segmentation Annotator with Character-level Visual Prompts
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.19917