Size-Invariant Attention Accuracy Metric for Image Captioning with High-Resolution Residual Attention.

Zongjian Zhang, Qiang Wu, Yang Wang, Fang Chen

Browse the full DICTA paper archive.