DPBridge: Latent Diffusion Bridge for Dense Prediction

Diffusion models have shown remarkable capabilities in modeling complex data distributions by transforming noise into structured data through stochastic processes. However, when applied to dense prediction tasks whose goal is to capture per-pixel relationships between RGB images and dense signal maps, starting the sampling process from an uninformative Gaussian noise often leads to inefficient sampling and long latency. To overcome these challenges, we propose DPBridge, a generative framework that establishes direct mapping between input RGB images and dense signal maps based on a tractable bridge process. Furthermore, we introduce finetuning strategies to leverage a pretrained large-scale image diffusion backbone, enjoying its rich visual prior knowledge to enable both efficient training and robust generalization. Experiments show that DPBridge achieves competitive performance compared to both feed-forward and diffusion-based approaches across various benchmarks, validating its effectiveness and adaptability.

@article{ji2025_2412.20506,
  title={ DPBridge: Latent Diffusion Bridge for Dense Prediction },
  author={ Haorui Ji and Taojun Lin and Hongdong Li },
  journal={arXiv preprint arXiv:2412.20506},
  year={ 2025 }
}