UniPercept：面向通用视觉感知的统一扩散模型
UniPercept: A Unified Diffusion Model for Generalizable Visual Perception
提出基于扩散模型的基础-适配器范式，用共享DiT骨干学习跨任务通用视觉表征，仅需少量样本和参数即可适配新任务，为视觉感知提供可复用的预训练-微调框架。
原文依据 a shared diffusion-based foundation model is trained to capture common and generalizable visual knowledge across diverse perception tasks
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论扩散模型在生成任务中展现出令人印象深刻的性能，证明了其捕捉详细结构和语义信息的能力。最近，这些能力已被扩展到视觉理解领域，研究利用扩散模型作为各种感知任务的骨干网络。然而，现有的基于扩散的感知模型通常局限于单一任务或一组固定的预定义任务，缺乏泛化到新任务的有效机制。为了克服这一限制，我们提出了一个基于DiT的统一感知框架UniPercept，它引入了一种新颖的基础-适配器范式用于通用视觉感知。在该框架中，一个共享的基于扩散的基础模型被训练以捕捉跨不同感知任务的常见且可泛化的视觉知识，并为每个单独任务集成任务特定的适配器。利用其卓越的泛化能力，基础模型可以通过轻量级适配器高效地适应新领域，仅需少至**1,000**个训练样本和不到**1%**的可训练参数。此外，UniPercept在各种感知任务上展现出强大性能，在大多数情况下优于最优的通用模型，并达到与专家模型相当的准确率。项目页面：https://VIPL-GENUN.github.io/Project-UniPercept。
英文原文摘要 1420 字符
Diffusion models have shown impressive performance in generative tasks, demonstrating their ability to capture detailed structural and semantic information. Recently, these capabilities have been extended to visual understanding, with studies employing diffusion models as the backbone for various perception tasks. However, existing diffusion-based perception models are generally restricted to a single task or a fixed set of predefined tasks, lacking an efficient mechanism to generalize to novel tasks. To overcome this limitation, we propose a unified DiT-based perception framework called UniPercept, which introduces a novel foundation-adapter paradigm for general visual perception. In this framework, a shared diffusion-based foundation model is trained to capture common and generalizable visual knowledge across diverse perception tasks, with task-specific adapters integrated for each individual task. Leveraging its superior generalization ability, the foundation model can be efficiently adapted to novel domains through lightweight adapters, requiring as few as **1,000** training samples and less than **1%** of trainable parameters. Furthermore, UniPercept demonstrates strong performance across various perception tasks, outperforming state-of-the-art generalist models in most cases and achieving accuracy comparable to specialist models. Project page: https://VIPL-GENUN.github.io/Project-UniPercept.