Embedding Font Impression Word Tags Based on Co-occurrence
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908503968317440 |
|---|---|
| author | Kubota, Yugo Uchida, Seiichi |
| author_facet | Kubota, Yugo Uchida, Seiichi |
| contents | Different font styles (i.e., font shapes) convey distinct impressions, indicating a close relationship between font shapes and word tags describing those impressions. This paper proposes a novel embedding method for impression tags that leverages these shape-impression relationships. For instance, our method assigns similar vectors to impression tags that frequently co-occur in order to represent impressions of fonts, whereas standard word embedding methods (e.g., BERT and CLIP) yield very different vectors. This property is particularly useful for impression-based font generation and font retrieval. Technically, we construct a graph whose nodes represent impression tags and whose edges encode co-occurrence relationships. Then, we apply spectral embedding to obtain the impression vectors for each tag. We compare our method with BERT and CLIP in qualitative and quantitative evaluations, demonstrating that our approach performs better in impression-guided font generation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_18825 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Embedding Font Impression Word Tags Based on Co-occurrence Kubota, Yugo Uchida, Seiichi Computer Vision and Pattern Recognition Different font styles (i.e., font shapes) convey distinct impressions, indicating a close relationship between font shapes and word tags describing those impressions. This paper proposes a novel embedding method for impression tags that leverages these shape-impression relationships. For instance, our method assigns similar vectors to impression tags that frequently co-occur in order to represent impressions of fonts, whereas standard word embedding methods (e.g., BERT and CLIP) yield very different vectors. This property is particularly useful for impression-based font generation and font retrieval. Technically, we construct a graph whose nodes represent impression tags and whose edges encode co-occurrence relationships. Then, we apply spectral embedding to obtain the impression vectors for each tag. We compare our method with BERT and CLIP in qualitative and quantitative evaluations, demonstrating that our approach performs better in impression-guided font generation. |
| title | Embedding Font Impression Word Tags Based on Co-occurrence |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2508.18825 |