Hierarchical Novelty Detection for Visual Object Recognition

Abstract

Deep neural networks have achieved impressive success in large-scale visual object recognition tasks with a predefined set of classes. However, recognizing objects of novel classes unseen during training still remains challenging. The problem of detecting such novel classes has been addressed in the literature, but most prior works have focused on providing simple binary or regressive decisions, e.g., the output would be "known," "novel," or corresponding confidence intervals. In this paper, we study more informative novelty detection schemes based on a hierarchical classification framework. For an object of a novel class, we aim for finding its closest super class in the hierarchical taxonomy of known classes. To this end, we propose two different approaches termed top-down and flatten methods, and their combination as well. The essential ingredients of our methods are confidence-calibrated classifiers, data relabeling, and the leave-one-out strategy for modeling novel classes under the hierarchical taxonomy. Furthermore, our method can generate a hierarchical embedding that leads to improved generalized zero-shot learning performance in combination with other commonly-used semantic embeddings.

Overcoming Catastrophic Forgetting With Unlabeled Data in the Wild

“…Therefore, we propose to sample an external dataset by a principled sampling strategy. To sample an effective external dataset from a large stream of unlabeled data, we propose to train a confidence-calibrated model [19,20] by utilizing irrelevant data as out-of-distribution (OOD) 2 samples. We show that unlabeled data from OOD should also be sampled for maintaining the model to be more confidence-calibrated.…”

A Vision-Based Two-Stage Framework for Inferring Physical Properties of the Terrain

“…Some GPU packages, such as Nvidia Jetson TX1 [18], are ideal for onboard processing. Algorithms based on deep learning have achieved excellent results in many fields of vision because of its powerful feature extraction ability [19][20][21].…”

Embedding Semantic Hierarchy in Discrete Optimal Transport for Risk Minimization

“…work structure or design a complicate inference logic that suffers from the effective optimization algorithms [6,7,8].…”

“…We compare the proposed model with classic and state-ofthe-art algorithms, including TSS [3], DARTS [17], RMGA [6], TDKL [7], CSMSE [8] and HSRM [4].…”