Novel Categories Discovery (NCD) aims to cluster novel data based on the class semantics of known classes using the open-world partial class space annotated dataset. As an alternative to the traditional pseudo-labeling-based approaches, we leverage the connection between the data sampling and the provided multinoulli (categorical) distribution of novel classes. We introduce constraints on individual and collective statistics of predicted novel class probabilities to implicitly achieve semantic-based clustering. More specifically, we align the class neuron activation distributions under Monte-Carlo sampling of novel classes in large batches by matching their empirical first-order (mean) and second-order (covariance) statistics with the multinoulli distribution of the labels while applying instance information constraints and prediction consistency under label-preserving augmentations. We then explore a directional statistics-based probability formation that learns the mixture of Von Mises-Fisher distribution of class labels in a unit hypersphere. We demonstrate the discriminative ability of our approach to realize semantic clustering of novel samples in image, video, and time-series modalities. We perform extensive ablation studies regarding data, networks, and framework components to provide better insights. Our approach maintains 94%, 93%, 85%, and 93% (approx.) classification accuracy in labeled data while achieving 90%, 84%, 72%, and 75% (approx.) clustering accuracy for novel categories in Cifar10, UCF101, MPSC-ARL, and SHAR datasets that match state-of-the-art approaches without any external clustering.

翻译：新类别发现（NCD）旨在利用开放世界部分类别空间标注数据集，基于已知类别的类语义对新颖数据进行聚类。作为传统伪标签方法的一种替代方案，我们利用数据采样与新颖类别提供的多项（类别）分布之间的联系。我们对预测的新颖类别概率的个体统计量和集体统计量引入约束，以隐式实现基于语义的聚类。具体而言，我们在对新颖类别进行蒙特卡洛采样的大批次中，通过将其实验一阶（均值）和二阶（协方差）统计量与标签的多项分布相匹配，同时应用实例信息约束和标签保持增强下的预测一致性，来对齐类别神经元激活分布。然后，我们探索一种基于方向统计的概率形成方法，该方法在单位超球面上学习类别标签的冯·米塞斯-费舍尔分布混合。我们展示了该方法在图像、视频和时间序列模态中实现新颖样本语义聚类的判别能力。我们针对数据、网络和框架组件进行了广泛的消融研究，以提供更深入的见解。我们的方法在Cifar10、UCF101、MPSC-ARL和SHAR数据集上，保持了约94%、93%、85%和93%的标注数据分类准确率，同时实现了约90%、84%、72%和75%的新颖类别聚类准确率，与现有最先进方法相当，且无需任何外部聚类。