TY - CHAP

T1 - Harvesting information from captions for weakly supervised semantic segmentation

AU - Sawatzky, Johann

AU - Banerjee, Debayan

AU - Gall, Juergen

N1 - Conference code: 17

PY - 2019/10

Y1 - 2019/10

N2 - Since acquiring pixel-wise annotations for training convolutional neural networks for semantic image segmentation is time-consuming, weakly supervised approaches that only require class tags have been proposed. In this work, we propose another form of supervision, namely image captions as they can be found on the Internet. These captions have two advantages. They do not require additional curation as it is the case for the clean class tags used by current weakly supervised approaches and they provide textual context for the classes present in an image. To leverage such textual context, we deploy a multi-modal network that learns a joint embedding of the visual representation of the image and the textual representation of the caption. The network estimates text activation maps (TAMs) for class names as well as compound concepts, i.e. combinations of nouns and their attributes. The TAMs of compound concepts describing classes of interest substantially improve the quality of the estimated class activation maps which are then used to train a network for semantic segmentation. We evaluate our method on the COCO dataset where it achieves state of the art results for weakly supervised image segmentation.

AB - Since acquiring pixel-wise annotations for training convolutional neural networks for semantic image segmentation is time-consuming, weakly supervised approaches that only require class tags have been proposed. In this work, we propose another form of supervision, namely image captions as they can be found on the Internet. These captions have two advantages. They do not require additional curation as it is the case for the clean class tags used by current weakly supervised approaches and they provide textual context for the classes present in an image. To leverage such textual context, we deploy a multi-modal network that learns a joint embedding of the visual representation of the image and the textual representation of the caption. The network estimates text activation maps (TAMs) for class names as well as compound concepts, i.e. combinations of nouns and their attributes. The TAMs of compound concepts describing classes of interest substantially improve the quality of the estimated class activation maps which are then used to train a network for semantic segmentation. We evaluate our method on the COCO dataset where it achieves state of the art results for weakly supervised image segmentation.

KW - Multimodal learning

KW - Semantic segmentation

KW - Weakly supervised learning

KW - Weakly supervised semantic segmentation

KW - Informatics

UR - http://www.scopus.com/inward/record.url?scp=85082499279&partnerID=8YFLogxK

U2 - 10.1109/ICCVW.2019.00549

DO - 10.1109/ICCVW.2019.00549

M3 - Article in conference proceedings

AN - SCOPUS:85082499279

SN - 978-1-7281-5024-6

T3 - IEEE International Conference on Computer Vision workshops

SP - 4481

EP - 4490

BT - 2019 International Conference on Computer Vision Workshops

PB - Institute of Electrical and Electronics Engineers Inc.

CY - Piscataway

T2 - 17th IEEE/CVF International Conference on Computer Vision Workshop - ICCVW 2019

Y2 - 27 October 2019 through 28 October 2019

ER -