BriefGPT.xyz
Jan, 2023
跨任务和领域传递良好特征的学习
Learning Good Features to Transfer Across Tasks and Domains
HTML
PDF
Pierluigi Zama Ramirez, Adriano Cardace, Luca De Luigi, Alessio Tonioni, Samuele Salti...
TL;DR
本文提出一种基于知识迁移和神经网络的方法,使得在新领域中,能够在有限或无额外监督的情况下利用特定领域中已学知识解决新任务,并且在特征空间约束和映射网络方面提出了一些策略,从而在合成到真实的语境下,利用单目深度估计和语义分割任务之间的知识迁移,获取了令人满意的结果。
Abstract
Availability of labelled data is the major obstacle to the deployment of
deep learning
algorithms for
computer vision
tasks in new domains. The fact that many frameworks adopted to solve different tasks share the
→