BriefGPT.xyz
Aug, 2018
使用选择性代理标签的FusionNet和AugmentedFlowNet用于未标记图像的训练
FusionNet and AugmentedFlowNet: Selective Proxy Ground Truth for Training on Unlabeled Images
HTML
PDF
Osama Makansi, Eddy Ilg, Thomas Brox
TL;DR
本研究介绍了如何使用合成数据集来增强预训练的卷积神经网络,从而提高其在实际领域中的性能,尤其在光流估计领域取得了最先进的成果。
Abstract
Recent work has shown that
convolutional neural networks
(CNNs) can be used to estimate
optical flow
with high quality and fast runtime. This makes them preferable for real-world applications. However, such netwo
→