AbstractDifferent from fine-tuning models pre-trained on a large-scale dataset of preset classes, class-incremental learning (CIL) aims to recognize novel classes over time without forgetting classes. However, given model will be challenged by test images with finer-grained e.g., basenji is at most recognized as dog. Such form new training set (i.e., support set) so that the incremental hoped q...

In this work, we propose the LoGID (Local and Global Incremental Learning for Dynamic Selection) framework, the main goal of which is to adapt hidden Markov model-based pattern recognition systems during both the generalization and learning phases. Given that the baseline system is composed of a pool of base classifiers, adaptation during generalization is performed through the dynamic selectio...

A fast SVM learning algorithm is proposed according to incremental learning and center convex hull operator. It is established on analyzing the relevance of support vector and convex hull from the angle of calculation geometry. The convex hull of current training samples is solved in the first place. Further, Euclidean distance elimination is applied to convex hull. Meanwhile, every time when t...

To solve the problem of tracking the trajectory of a moving object and learning a deep compact image representation in the complex environment, a novel robust incremental deep learning tracker is presented under the particle filter framework. The incremental deep classification neural network was composed of stacked denoising autoencoder, incremental feature learning and support vector machine ...

Class-incremental learning (CIL) aims to train a classification model while the number of classes increases phase-by-phase. An inherent challenge CIL is stability-plasticity tradeoff, i.e., models should keep stable retain old knowledge and plastic absorb new knowledge. However, none existing can achieve optimal tradeoff in different data-receiving settings—where typically training-from-half (T...