AviaTAD-LGH: A Multi-Task Spatio-Temporal Action Detector with Lightweight Gradient Harmonization for Real-Time Avian Behavior Monitoring

Fine-grained spatio-temporal action detection in continuous, unconstrained field videos remains a formidable challenge due to severe background clutter, high inter-class similarity, and the scarcity of domain-specific benchmarks. To address these limitations, we first introduce a large-scale Winteri...

Popoln opis

Bibliografske podrobnosti

Principais autores:	Zihui Xie, Haifang Jian, Wenhui Yang, Mengdi Fu, Wanting Peng, Markus Peter Eichhorn, Ramiro Daniel Crego, Xin Ning, Jun Du, Hongchang Wang
Format:	Artigo
Jezik:	Inglês
Izdano:	MDPI AG 2026-03-01
Serija:	Sensors
Teme:	video sensing deep learning spatio-temporal action detection multi-task learning ecological monitoring
Online dostop:	https://www.mdpi.com/1424-8220/26/7/2088
Oznake:	Označite Brez oznak, prvi označite!