1 paper
Seong-Jin Park, Ki-Sang Hong
This paper proposes a new framework for semantic segmentation of objects in videos. We address the label inconsistency problem of deep convolutional neural networks (DCNNs) by expl…