arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2302.10326cs.CVcs.LG
收藏

基于扩散修复的无监督分布外检测

Unsupervised Out-of-Distribution Detection with Diffusion Inpainting

Cornell University(康奈尔大学)

Zhenzhen Liu, Jin Peng Zhou, Yufan Wang, Kilian Q. Weinberger

AI总结:

提出无监督OOD检测方法LMD,利用扩散模型先破坏图像再映射回域内流形,并依据映射后与原始流形的距离识别域外样本,在多数据集上表现具有竞争力。

摘要:

无监督分布外检测(OOD)旨在仅通过学习未标注的域内数据来识别域外数据。我们针对该任务提出了一种新颖方法——Lift, Map, Detect(LMD),它利用了扩散模型的最新进展。扩散模型是生成模型的一种,其核心是学习一个迭代去噪过程,将含噪图像逐渐映射到更接近其训练流形的位置。LMD将这一直觉用于OOD检测。具体而言,LMD通过破坏图像将其抬离原始流形,再用扩散模型将其映射向域内流形。对于域外图像,映射后的图像会与原始流形保持较大距离,LMD据此将其识别为OOD。大量实验表明,LMD在多种多样的数据集上均取得了具有竞争力的性能。代码可在 https://github.com/zhenzhel/lift_map_detect 获取。

英文摘要:

Unsupervised out-of-distribution detection (OOD) seeks to identify out-of-domain data by learning only from unlabeled in-domain data. We present a novel approach for this task - Lift, Map, Detect (LMD) - that leverages recent advancement in diffusion models. Diffusion models are one type of generative models. At their core, they learn an iterative denoising process that gradually maps a noisy image closer to their training manifolds. LMD leverages this intuition for OOD detection. Specifically, LMD lifts an image off its original manifold by corrupting it, and maps it towards the in-domain manifold with a diffusion model. For an out-of-domain image, the mapped image would have a large distance away from its original manifold, and LMD would identify it as OOD accordingly. We show through extensive experiments that LMD achieves competitive performance across a broad variety of datasets. Code can be found at https://github.com/zhenzhel/lift_map_detect.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑