论文链接:[2304.14310] Incremental Generalized Category Discovery (arxiv.org) 代码链接:DTennant/Incremental-Generalized-Category-Discovery (github.com) 摘要 我们探讨了增量通用类别发现(IGCD)问题。这是一个极具挑战性的增量分类学习问题,其目标是开发出能够正确分类以前所见类别图像的模型,同时发现新的类别。学习...
在四个基准数据集上进行的实验表明,所提出的方法优于几种竞争方法,同时所需的参数也更少。 【2】 MetaGCD: Learning to Continually Learn in Generalized Category Discovery ICCV 2023 北京交通大学 多伦多大学 MetaGCD: Learning to Continually Learn in Generalized Category Discovery | PDF (arxiv.org) In th...
Proxy Anchor-based Unsupervised Learning for Continuous Generalized Category Discovery(ICCV 2023)[paper] [XLDA] XLDA: Linear Discriminant Analysis for Scaling Continual Learning to Extreme Classification at the Edge[ICML 2023][paper] [CLR] CLR: Channel-wise Lightweight Reprogramming for Continual Learni...
Full size image In Fig.8, we can see that common identifying feature such as name is suppressed at the very beginning. Then, other features which could be used to identify a person in a dataset are generalized by varying the value of K at a different time. In our case, we have used ...
FIG. 3 is a diagram representing a generalized constraint matrix; FIG. 4 is another exemplary small constraint matrix; FIG. 5 is a small exemplary portion of a mask; FIG. 6 is an exemplary application of the FIG. 4 constraint matrix to the FIG. 5 mask; ...
Wang, E., Peng, Z., Xie, Z., Yang, F., Liu, X., Cheng, M.M.: Unlocking the multi-modal potential of clip for generalized category discovery (2024).https://arxiv.org/abs/2403.09974 Wang, F.Y., Zhou, D.W., Ye, H.J., Zhan, D.C.: FOSTER: feature boosting and compression...
(unknown or specifically known) based on the uncertainty of the classifier'sprediction; (b) For the first time, we introduce generalized category discovery (GCD) techniques in OWR to assist humans in grouping unlabeled data; (c) For the smoothexecutionof IL and GCD, we retain an equal ...