Rebalancing in Supervised Contrastive Learning for Long-Tailed Visual Recognition

In real-world visual recognition tasks, long-tailed distribution is a pervasive challenge, where the extreme class imbalance severely limits the representation learning capability of deep models. Although supervised learning has demonstrated certain potential in long-tailed visual recognition, these...

Celý popis

Podrobná bibliografie

Hlavní autoři:	Jiahui Lv, Jun Lei, Jun Zhang, Chao Chen, Shuohao Li
Médium:	Artigo
Jazyk:	Inglês
Vydáno:	MDPI AG 2025-08-01
Edice:	Big Data and Cognitive Computing
Témata:	long-tailed visual recognition contrastive learning representation learning feature balancing
On-line přístup:	https://www.mdpi.com/2504-2289/9/8/204
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!