Seeing unseen: Discover novel biomedical concepts via geometry-constrained probabilistic modeling

J Fan, D Liu, H Chang, H Huang… - Proceedings of the …, 2024‏ - openaccess.thecvf.com
Abstract Machine learning holds tremendous promise for transforming the fundamental
practice of scientific discovery by virtue of its data-driven nature. With the ever-increasing …

Revisiting Unsupervised Temporal Action Localization: The Primacy of High-Quality Actionness and Pseudolabels

H Jiang, H Tang, M Yan, J Zhang, M Xu, Y Hu… - Proceedings of the …, 2024‏ - dl.acm.org
Recently, temporal action localization (TAL) methods, especially the weakly-supervised and
unsupervised ones, have become a hot research topic. Existing unsupervised methods …

Progressive transformation learning for leveraging virtual images in training

YT Shen, H Lee, H Kwon… - Proceedings of the …, 2023‏ - openaccess.thecvf.com
To effectively interrogate UAV-based images for detecting objects of interest, such as
humans, it is essential to acquire large-scale UAV-based datasets that include human …

A survey on open-set image recognition

J Sun, Q Dong - arxiv preprint arxiv:2312.15571, 2023‏ - arxiv.org
Open-set image recognition (OSR) aims to both classify known-class samples and identify
unknown-class samples in the testing set, which supports robust classifiers in many realistic …

Open-Set Text Recognition Implementations (III): Open-set Predictor

XC Yin, C Yang, C Liu - Open-Set Text Recognition: Concepts, Framework …, 2024‏ - Springer
This chapter discusses the approaches of the representation-prototype matching process,
which is used to recognize or reject the corresponding samples in question. For each query …

Learning for transductive threshold calibration in open-world recognition

Q Zhang, D An, T **ao, T He, Q Tang… - Proceedings of the …, 2024‏ - openaccess.thecvf.com
In deep metric learning for visual recognition the calibration of distance thresholds is crucial
for achieving desired model performance in the true positive rates (TPR) or true negative …

Open-world dynamic prompt and continual visual representation learning

Y Kim, J Fang, Q Zhang, Z Cai, Y Shen… - … on Computer Vision, 2024‏ - Springer
The open world is inherently dynamic, characterized by ever-evolving concepts and
distributions. Continual learning (CL) in this dynamic open-world environment presents a …