Trustworthy SR: Resolving Ambiguity in Image Super-resolution via Diffusion Models and Human Feedback

Documenti analoghi

• Training Generative Image Super-Resolution Models by Wavelet-Domain Losses Enables Better Control of Artifacts
  di: Korkmaz, Cansu, et al.
  Pubblicazione: (2024)
• Training Transformer Models by Wavelet Losses Improves Quantitative and Visual Performance in Single Image Super-Resolution
  di: Korkmaz, Cansu, et al.
  Pubblicazione: (2024)
• DiMoSR: Feature Modulation via Multi-Branch Dilated Convolutions for Efficient Image Super-Resolution
  di: Yilmaz, M. Akin, et al.
  Pubblicazione: (2025)
• Leveraging Vision-Language Models to Select Trustworthy Super-Resolution Samples Generated by Diffusion Models
  di: Korkmaz, Cansu, et al.
  Pubblicazione: (2025)
• AdaptSR: Low-Rank Adaptation for Efficient and Scalable Real-World Super-Resolution
  di: Korkmaz, Cansu, et al.
  Pubblicazione: (2025)