Integrating Amortized Inference with Diffusion Models for Learning Clean Distribution from Corrupted Images

Documenti analoghi

• An Expectation-Maximization Algorithm for Training Clean Diffusion Models from Corrupted Observations
  di: Bai, Weimin, et al.
  Pubblicazione: (2024)
• Vision-Language Models as Differentiable Semantic and Spatial Rewards for Text-to-3D Generation
  di: Bai, Weimin, et al.
  Pubblicazione: (2025)
• Unbiased Diffusion Variational Inversion via Principled Posterior Matching
  di: Bai, Weimin, et al.
  Pubblicazione: (2026)
• Dive3D: Diverse Distillation-based Text-to-3D Generation via Score Implicit Matching
  di: Bai, Weimin, et al.
  Pubblicazione: (2025)
• Uni-Instruct: One-step Diffusion Model through Unified Diffusion Divergence Instruction
  di: Wang, Yifei, et al.
  Pubblicazione: (2025)