Learning to Exploit the Prior Network Knowledge for Weakly-Supervised Semantic Segmentation

13 Apr 2018Carolina Redondo-CabreraMarcos Baptista-RíosRoberto J. López-Sastre

Training a Convolutional Neural Network (CNN) for semantic segmentation typically requires to collect a large amount of accurate pixel-level annotations, a hard and expensive task. In contrast, simple image tags are easier to gather... (read more)

PDF Abstract

Evaluation results from the paper


  Submit results from this paper to get state-of-the-art GitHub badges and help community compare results to other papers.