Fudan's TextRIRO enables scene text editing with self-paired real data only
Fudan-University · hf · 2026-10-22
TextRIRO from Fudan University and Tencent WeChat Vision challenges the reliance on synthetic paired data in scene text editing. By disrupting each real image (Crop, Shuffle, Mask) to create self-paired supervision and training a conditional diffusion model, it improves realism, reduces artifacts and hallucinations, and benefits downstream text recognition — using only real data.
More from Multimodal
- fal employee shows Claude + fal generating a game in 60 minutes — OdinLovis · 2026-10-22
- Indie dev uses AI to 1:1 recreate his own RV for his fourth game — ezshine · 2026-10-22
- User Shows Off Cinematic AI Video Made with Seedance 2.5 — SimplyAnnisa · 2026-10-22
- StreamEval benchmarks streaming video generation: best model scores just 36.90 on task execution — StreamEval Team · 2026-10-22
- Spline launches ChatGPT integration to create and edit interactive 3D scenes in chat — algo_diver · 2026-10-22
- 8-step and 4-step PTD acceleration LoRAs released for MiniMax-H3 Ref2VA — Apprehensive_Sky892 · 2026-10-22