Adding RGB surface-normal masks to synthetic scene-text data improves perspective handling for a MOSTEL-based text editor, though several reported gains are inconsistent with the tables.
Textdiffuser: Diffusion models as text painters
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Syn3DTxt: Embedding 3D Cues for Scene Text Generation
Adding RGB surface-normal masks to synthetic scene-text data improves perspective handling for a MOSTEL-based text editor, though several reported gains are inconsistent with the tables.