English

Modulating Bottom-Up and Top-Down Visual Processing via Language-Conditional Filters

Computer Vision and Pattern Recognition 2022-06-24 v3 Computation and Language Machine Learning

Abstract

How to best integrate linguistic and perceptual processing in multi-modal tasks that involve language and vision is an important open problem. In this work, we argue that the common practice of using language in a top-down manner, to direct visual attention over high-level visual features, may not be optimal. We hypothesize that the use of language to also condition the bottom-up processing from pixels to high-level features can provide benefits to the overall performance. To support our claim, we propose a U-Net-based model and perform experiments on two language-vision dense-prediction tasks: referring expression segmentation and language-guided image colorization. We compare results where either one or both of the top-down and bottom-up visual branches are conditioned on language. Our experiments reveal that using language to control the filters for bottom-up visual processing in addition to top-down attention leads to better results on both tasks and achieves competitive performance. Our linguistic analysis suggests that bottom-up conditioning improves segmentation of objects especially when input text refers to low-level visual concepts. Code is available at https://github.com/ilkerkesen/bvpr.

Keywords

Cite

@article{arxiv.2003.12739,
  title  = {Modulating Bottom-Up and Top-Down Visual Processing via Language-Conditional Filters},
  author = {İlker Kesen and Ozan Arkan Can and Erkut Erdem and Aykut Erdem and Deniz Yuret},
  journal= {arXiv preprint arXiv:2003.12739},
  year   = {2022}
}

Comments

13 pages, 6 figures, 6 tables. Appeared in MULA Workshop at CVPR 2022

R2 v1 2026-06-23T14:30:05.962Z