Few-shot object counting (FSC) aims to estimate counts for previously unseen (novel) categories given only a few exemplars, leveraging knowledge learned on seen (base) categories for generalization. However, most existing methods rely on unidirectional matching and underuse scene context, limiting the effective use of...
Jing-Hui He, Bo Liu, Fan Cao et al.· Neural Networks· 0 citations
This work designs a class-aware weighted similarity strategy to accurately impute missing PI and dynamically assign confidence weights to the downstream SVM+ classifier and confirms EPPL as a highly robust solution for leveraging incomplete privileged information in complex environments.
Bo Liu, Yun-Long Sun, Yan-Shan Xiao et al.· Neural computing & applicati...· 0 citations
Improved robustness in noisy and imbalanced settings is indicated, and the reduced performance variance across repeated trials indicates improved robustness in noisy and imbalanced settings.
Bo Liu, Jia-Jun Chen, Yanshan Xiao et al.· International Journal of Mac...· 0 citations
SMVPU first extracts reliable negative samples from the unlabeled set and assigns similarity-weighted values to the remaining unlabeled instances, and then incorporates multi-view representations to enhance feature compatibility and discriminability.
Bo Liu, Wentao Li, Fan Cao et al.· Applied intelligence (Boston...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
