Rethinking few-shot class-incremental learning with open-set hypothesis in hyperbolic geometry

X Dengxiong, Y Kong - … of the IEEE/CVF Winter Conference …, 2023 - openaccess.thecvf.com

Different from the open set recognition, generalized open set recognition learns the most
similar known classes for unseen samples using known classes samples and side …

被引用次数：7 相关文章所有 3 个版本

Modeling Inner-and Cross-Task Contrastive Relations for Continual Image Classification

Y Luo, R Cong, X Liu, HHS Ip… - IEEE Transactions on …, 2024 - ieeexplore.ieee.org

Existing continual image classification methods demonstrate that samples from all
sequences of continual classification tasks contain common (task-invariant) features and …

被引用次数：2 相关文章所有 2 个版本

[PDF] arxiv.org

Taxonomy-Aware Continual Semantic Segmentation in Hyperbolic Spaces for Open-World Perception

J Hindel, D Cattaneo, A Valada - IEEE Robotics and …, 2024 - ieeexplore.ieee.org

Semantic segmentation models are typically trained on a fixed set of classes, limiting their
applicability in open-world scenarios. Class-incremental semantic segmentation aims to …

Controllable Relation Disentanglement for Few-Shot Class-Incremental Learning

Y Zhou, R Hong, Y Guo, L Liu, S Hao… - arXiv preprint arXiv …, 2024 - arxiv.org

In this paper, we propose to tackle Few-Shot Class-Incremental Learning (FSCIL) from a
new perspective, ie, relation disentanglement, which means enhancing FSCIL via …

Mitigating Label Noise using Prompt-Based Hyperbolic Meta-Learning in Open-Set Domain Generalization

K Peng, D Wen, SM Saquib, Y Chen, J Zheng… - arXiv preprint arXiv …, 2024 - arxiv.org

Open-Set Domain Generalization (OSDG) is a challenging task requiring models to
accurately predict familiar categories while minimizing confidence for unknown categories to …

[PDF] arxiv.org

Taming self-supervised learning for presentation attack detection: De-folding and de-mixing

Z Kong, W Zhang, F Liu, W Luo, H Liu… - … on Neural Networks …, 2023 - ieeexplore.ieee.org

Biometric systems are vulnerable to presentation attacks (PAs) performed using various PA
instruments (PAIs). Even though there are numerous PA detection (PAD) techniques based …

被引用次数：4 相关文章所有 5 个版本

[PDF] arxiv.org

OpenSlot: Mixed Open-set Recognition with Object-centric Learning

X Yin, F Pan, G An, Y Huo, Z Xie, SE Yoon - arXiv preprint arXiv …, 2024 - arxiv.org

Existing open-set recognition (OSR) studies typically assume that each image contains only
one class label, and the unknown test set (negative) has a disjoint label space from the …

Sharpness-Aware Geometric Defense for Robust Out-Of-Distribution Detection

JL Li, MC Chang, WC Chen - openreview.net

Out-of-distribution (OOD) detection ensures safe and reliable model deployment.
Contemporary OOD algorithms using geometry projection can detect OOD or adversarial …