Lotus: Diffusion-Based Visual Foundation Model for High-Quality Dense Prediction

LOTUS is a state-of-the-art visual foundation model presented in arXiv 2024, using diffusion techniques for zero-shot dense geometry prediction tasks. Its novel approach and fine-tuning protocol allow it to excel in depth and normal estimation benchmarks with minimal training data, outperforming other methods in quality and efficiency.

• LOTUS is a diffusion-based foundation model.
• Achieves SoTA in zero-shot depth estimation.
• Outperforms in zero-shot surface normal estimation.
• Employs a single-step diffusion process.
• Designed for high efficiency and fine-grained prediction.