Leveraging Text Signed Distance Function Map for Boundary-Aware Guidance in Scene Text Segmentation
Scene text segmentation is to predict pixel-wise text regions from an image, enabling in-image text editing or removal. One of the primary challenges is to remove noises including non-text regions and predict intricate text boundaries. To deal with that, traditional approaches utilize a text detecti...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10981835/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850143643277459456 |
|---|---|
| author | Ho Jun Kim Hak Gu Kim |
| author_facet | Ho Jun Kim Hak Gu Kim |
| author_sort | Ho Jun Kim |
| collection | DOAJ |
| description | Scene text segmentation is to predict pixel-wise text regions from an image, enabling in-image text editing or removal. One of the primary challenges is to remove noises including non-text regions and predict intricate text boundaries. To deal with that, traditional approaches utilize a text detection or recognition module explicitly. However, they are likely to highlight noise around the text. Because they did not sufficiently consider the boundaries of text, they fail to accurately predict the fine details of text. In this paper, we introduce leveraging text signed distance function (SDF) map, which encodes distance information from text boundaries, in scene text segmentation to explicitly provide text boundary information. By spatial cross attention mechanism, we encode the text-attended feature from the text SDF map. Then, both visual and text-attended features are utilized to decode the text segmentation map. Our approach not only mitigates confusion between text and complex backgrounds by eliminating false positives such as logos and texture blobs located far from the text, but also effectively captures fine details of complex text patterns by leveraging text boundary information. Extensive experiments demonstrate that leveraging text SDF map in scene text segmentation provides superior performances on various scene text segmentation datasets. |
| format | Article |
| id | doaj-art-fab1ee4e416146848effd554cd2ddeb4 |
| institution | OA Journals |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-fab1ee4e416146848effd554cd2ddeb42025-08-20T02:28:37ZengIEEEIEEE Access2169-35362025-01-0113788507886310.1109/ACCESS.2025.356604510981835Leveraging Text Signed Distance Function Map for Boundary-Aware Guidance in Scene Text SegmentationHo Jun Kim0https://orcid.org/0000-0002-3086-4809Hak Gu Kim1https://orcid.org/0000-0003-2137-934XDepartment of Image Science and Arts, GSAIM, Chung-Ang University, Seoul, South KoreaDepartment of Metaverse Convergence, GSAIM, Chung-Ang University, Seoul, South KoreaScene text segmentation is to predict pixel-wise text regions from an image, enabling in-image text editing or removal. One of the primary challenges is to remove noises including non-text regions and predict intricate text boundaries. To deal with that, traditional approaches utilize a text detection or recognition module explicitly. However, they are likely to highlight noise around the text. Because they did not sufficiently consider the boundaries of text, they fail to accurately predict the fine details of text. In this paper, we introduce leveraging text signed distance function (SDF) map, which encodes distance information from text boundaries, in scene text segmentation to explicitly provide text boundary information. By spatial cross attention mechanism, we encode the text-attended feature from the text SDF map. Then, both visual and text-attended features are utilized to decode the text segmentation map. Our approach not only mitigates confusion between text and complex backgrounds by eliminating false positives such as logos and texture blobs located far from the text, but also effectively captures fine details of complex text patterns by leveraging text boundary information. Extensive experiments demonstrate that leveraging text SDF map in scene text segmentation provides superior performances on various scene text segmentation datasets.https://ieeexplore.ieee.org/document/10981835/Scene text segmentationsigned distance functionoptical character recognition |
| spellingShingle | Ho Jun Kim Hak Gu Kim Leveraging Text Signed Distance Function Map for Boundary-Aware Guidance in Scene Text Segmentation IEEE Access Scene text segmentation signed distance function optical character recognition |
| title | Leveraging Text Signed Distance Function Map for Boundary-Aware Guidance in Scene Text Segmentation |
| title_full | Leveraging Text Signed Distance Function Map for Boundary-Aware Guidance in Scene Text Segmentation |
| title_fullStr | Leveraging Text Signed Distance Function Map for Boundary-Aware Guidance in Scene Text Segmentation |
| title_full_unstemmed | Leveraging Text Signed Distance Function Map for Boundary-Aware Guidance in Scene Text Segmentation |
| title_short | Leveraging Text Signed Distance Function Map for Boundary-Aware Guidance in Scene Text Segmentation |
| title_sort | leveraging text signed distance function map for boundary aware guidance in scene text segmentation |
| topic | Scene text segmentation signed distance function optical character recognition |
| url | https://ieeexplore.ieee.org/document/10981835/ |
| work_keys_str_mv | AT hojunkim leveragingtextsigneddistancefunctionmapforboundaryawareguidanceinscenetextsegmentation AT hakgukim leveragingtextsigneddistancefunctionmapforboundaryawareguidanceinscenetextsegmentation |