FALIP: Visual Prompt as Foveal Attention Boosts CLIP Zero-Shot Performance

Documenti analoghi

• Mitigating Hallucination in Visual-Language Models via Re-Balancing Contrastive Decoding
  di: Liang, Xiaoyu, et al.
  Pubblicazione: (2024)
• TeG-DG: Textually Guided Domain Generalization for Face Anti-Spoofing
  di: Mu, Lianrui, et al.
  Pubblicazione: (2023)
• Training-Free Multi-View Extension of IC-Light for Textual Position-Aware Scene Relighting
  di: Ye, Jiangnan, et al.
  Pubblicazione: (2025)
• No Pixel Left Behind: A Detail-Preserving Architecture for Robust High-Resolution AI-Generated Image Detection
  di: Mu, Lianrui, et al.
  Pubblicazione: (2025)
• Identity-Preserving Pose-Guided Character Animation via Facial Landmarks Transformation
  di: Mu, Lianrui, et al.
  Pubblicazione: (2024)