It is commonly assumed that language refers to high-level visual concepts while leaving low-level visual processing unaffected. This view dominates the current literature in computational models for language-vision tasks, where visual and linguistic input are mostly processed independently before being fused into a single representation. In this paper, we deviate from this classic pipeline and propose to modulate the \emph{entire visual processing} by linguistic input. Specifically, we condition the batch normalization parameters of a pretrained residual network (ResNet) on a language embedding. This approach, which we call MOdulated RESnet (\MRN), significantly improves strong baselines on two visual question answering tasks. Our ablation study shows that modulating from the early stages of the visual processing is beneficial.
@article{arxiv.1707.00683,
title = {Modulating early visual processing by language},
author = {Harm de Vries and Florian Strub and Jérémie Mary and Hugo Larochelle and Olivier Pietquin and Aaron Courville},
journal= {arXiv preprint arXiv:1707.00683},
year = {2017}
}
Comments
Advances in Neural Information Processing Systems 30 (NIPS 2017)