Open-World Semi-Supervised Learning(2020arXiv) === 这篇文章投ICLR2021没中,openreview链接: 作者在这篇文章中也提到了unlabeled set中可能出现novel class的情况,并且基于此场景提出了一种自动识别novel class的算法,取名为ORCA (stands for Open-woRld with unCertainty based Adaptive margin)。 作者首先通过自监...
Therefore, this paper considers a more realistic and widespread paradigm in which the labeled and unlabeled data come from the mismatched distribution, dubbed as Open-Set Semi-Supervised Learning (OS-SSL). Specifically, unlabeled data contains out of distribution (OOD) samples, which are samples ...
However, in practice, unlabeled data can contain categories unseen in the labeled set, i.e., outliers, which can significantly harm the performance of SSL algorithms. To address this problem, we propose a novel Open-set Semi-Supervised Learning (OSSL) approach called OpenMatch...
作者根据target domain是否有label,把问题分成了unsupervised和semi-supervised domain adaptation。然后分开解决。空间变换这一步是共同的。 Unsupervised domain adaptation 作者用xct来标识,target domain中的第t个样本是否被标记为类别c,xct∈{0,1}。同时,因为是个open set,所以,引入一个ot来标识第t个样本是否为未知...
Semi-supervised learning (SSL) methods assume that labeled data, unlabeled data and test data are from the same distribution. Open-set semi-supervised learning (Open-set SSL) considers a more practical scenario, where unlabeled data and test data contain new categories (outliers) not observed in...
几篇论文实现代码:《OpenMatch: Open-set Consistency Regularization for Semi-supervised Learning with Outliers》(NeurIPS 2021) GitHub:https:// github.com/VisionLearningGroup/OP_Match [fig8] 《Learn...
Supervised and semi-supervised learning methods have been traditionally designed for the closed-world setting based on the assumption that unlabeled test data contains only classes previously encountered in the labeled training data. However, the real world is inherently open and dynamic, and thus novel...
Open World Object Detection is a computer vision problem where a model is tasked to: 1) identify objects that have not been introduced to it as `unknown', without explicit supervision to do so, and 2) incrementally learn these identified unknown categori
Semi-supervised learning (SSL) is one of the main approaches to address the high cost of manual annotation in supervised learning. In recent years, SSL methods have effectively utilized consistency regularization on unlabeled data to improve performance while leveraging a small portion of labeled data...
因此,作者提出了一种更符合实际的研究问题——开集半监督目标检测(Open-set Semi-supervised Object Detection,OSSOD)。如图1(a),标注数据与一般的SSOD保持一致,只有分布内(In-distribution, OOD) 的类别。而无标注的数据里ID和OOD的类别的物体同时存在。由于OOD数据的干扰,经过实验发现,一般的SSOD算法性能出现下降...