Tag: contrastive loss

A coarse-to-fine two-stage deep learning framework combining YOLOv8m and MobileNetV3 achieves higher detection accuracy for electrical insulator defects in aerial ...

Researchers developed a unified spatio-temporal graph attention transformer framework that achieved 85.9 percent accuracy on DEAP and 91.3 percent accuracy ...

A new hybrid model called SST-Align improves topological network alignment by combining graphlet-based optimization with self-supervised graph convolutional networks.

Researchers have combined convolutional neural networks and vision transformers with a three-loss training strategy to significantly improve content-based image retrieval ...