This paper considers the problem of unsupervised person re-identification (re-ID), which aims to learn discriminative models with unlabeled data. One popular method is to obtain pseudo-label by clustering and use them to optimize the model. Although this kind of approach has shown promising accuracy, it is hampered by 1) noisy labels produced by clustering and 2) feature variations caused by camera shift. The former will lead to incorrect optimization and thus hinders the model accuracy. The latter will result in assigning the intra-class samples of different cameras to different pseudo-label, making the model sensitive to camera variations. In this paper, we propose a unified framework to solve both problems. Concretely, we propose a Dynamic and Symmetric Cross-Entropy loss (DSCE) to deal with noisy samples and a camera-aware meta-learning algorithm (MetaCam) to adapt camera shift. DSCE can alleviate the negative effects of noisy samples and accommodate the change of clusters after each clustering step. MetaCam simulates cross-camera constraint by splitting the training data into meta-train and meta-test based on camera IDs. With the interacted gradient from meta-train and meta-test, the model is enforced to learn camera-invariant features. Extensive experiments on three re-ID benchmarks show the effectiveness and the complementary of the proposed DSCE and MetaCam. Our method outperforms the state-of-the-art methods on both fully unsupervised re-ID and unsupervised domain adaptive re-ID.
翻译:本文探讨了无人监督的人重新认同(Re-ID)的问题,目的是学习带有未贴标签数据的歧视模型(re-ID),目的是学习使用未贴标签的数据来学习歧视性模型。一种流行的方法是通过集群获得假标签,并用它们优化模型。虽然这种方法显示了很有希望的准确性,但受到以下因素的阻碍:1)通过集群产生的噪音标签和2)由于摄像头变换而产生的特征变化;前者将导致不正确的优化,从而妨碍模型的准确性。后者将导致将不同相机的分类内样本分配到不同的假标签,从而使模型对相机的变化敏感。在本文中,我们提出了一个解决这两个问题的统一框架。具体地说,我们提出了一个动态和对称跨英特效损失(DSCE)框架,以处理噪音样品和照相机变换换换的摄像学元方法(Metecam) 。DSCE(Metrolemental)可以减轻噪音样品的消极影响,并在每次组合步骤后适应集群的变化。MetCam模拟了交叉摄像限制,将培训数据分解成以相机的元数据和元测试,以相机代码码码码码码码码码码码码为非代的不全。我们SDCD-DErearetretread-retradrol-read-retradreal-read-read-read-read-read-regleglegy-remodalma-de-remod-realma-remod-realma-remod-remod-remod-realmad-remotralegy-remod-reamental-real-real-real-real-real-real-real-real-real-real-real-real-read-real-real-real-lad-readalmasal-real-reabal-re-re-real-real-real-real-real-real-real-real-re-re-re-real-re-real-real-real-real-real-real-re-real-real-real-real-