Skip to Main content Skip to Navigation
Conference papers

Modulating early visual processing by language

Abstract : 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 inputs 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 entire visual processing by a linguistic input. Specifically, we introduce Conditional Batch Normalization (CBN) as an efficient mechanism to modulate convolutional feature maps by a linguistic embedding. We apply CBN to a pre-trained Residual Network (ResNet), leading to the MODulatEd ResNet (MODERN) architecture, and show that this significantly improves strong baselines on two visual question answering tasks. Our ablation study confirms that modulating from the early stages of the visual processing is beneficial.
Complete list of metadata

Cited literature [28 references]  Display  Hide  Download
Contributor : Florian Strub Connect in order to contact the contributor
Submitted on : Monday, November 27, 2017 - 3:06:06 AM
Last modification on : Friday, January 21, 2022 - 3:13:03 AM


Files produced by the author(s)


  • HAL Id : hal-01648683, version 1
  • ARXIV : 1707.00683


Harm de Vries, Florian Strub, Jérémie Mary, Hugo Larochelle, Olivier Pietquin, et al.. Modulating early visual processing by language. NIPS 2017 - Conference on Neural Information Processing Systems, Dec 2017, Long Beach, United States. pp.1-14. ⟨hal-01648683⟩



Les métriques sont temporairement indisponibles