ASTRA: Enhancing Multi-Subject Generation with Retrieval-Augmented Pose Guidance and Disentangled Position Embedding

Documenti analoghi

• ASTRA: Let Arbitrary Subjects Transform in Video Editing
  di: Shen, Fei, et al.
  Pubblicazione: (2025)
• DisPose: Disentangling Pose Guidance for Controllable Human Image Animation
  di: Li, Hongxiang, et al.
  Pubblicazione: (2024)
• MAGREF: Masked Guidance for Any-Reference Video Generation with Subject Disentanglement
  di: Deng, Yufan, et al.
  Pubblicazione: (2025)
• Are We on the Right Way for Assessing Document Retrieval-Augmented Generation?
  di: Shen, Wenxuan, et al.
  Pubblicazione: (2025)
• MOSAIC: Multi-Subject Personalized Generation via Correspondence-Aware Alignment and Disentanglement
  di: She, Dong, et al.
  Pubblicazione: (2025)