English

Diff-2-in-1: Bridging Generation and Dense Perception with Diffusion Models

Computer Vision and Pattern Recognition 2025-12-18 v1 Machine Learning Robotics

Abstract

Beyond high-fidelity image synthesis, diffusion models have recently exhibited promising results in dense visual perception tasks. However, most existing work treats diffusion models as a standalone component for perception tasks, employing them either solely for off-the-shelf data augmentation or as mere feature extractors. In contrast to these isolated and thus sub-optimal efforts, we introduce a unified, versatile, diffusion-based framework, Diff-2-in-1, that can simultaneously handle both multi-modal data generation and dense visual perception, through a unique exploitation of the diffusion-denoising process. Within this framework, we further enhance discriminative visual perception via multi-modal generation, by utilizing the denoising network to create multi-modal data that mirror the distribution of the original training set. Importantly, Diff-2-in-1 optimizes the utilization of the created diverse and faithful data by leveraging a novel self-improving learning mechanism. Comprehensive experimental evaluations validate the effectiveness of our framework, showcasing consistent performance improvements across various discriminative backbones and high-quality multi-modal data generation characterized by both realism and usefulness.

Keywords

Cite

@article{arxiv.2411.05005,
  title  = {Diff-2-in-1: Bridging Generation and Dense Perception with Diffusion Models},
  author = {Shuhong Zheng and Zhipeng Bao and Ruoyu Zhao and Martial Hebert and Yu-Xiong Wang},
  journal= {arXiv preprint arXiv:2411.05005},
  year   = {2025}
}

Comments

26 pages, 14 figures

R2 v1 2026-06-28T19:52:07.835Z