Show simple item record

dc.contributor.author
Zhang, Meijie
dc.contributor.author
Li, Jianwu
dc.contributor.author
Zhou, Tianfei
dc.date.accessioned
2023-04-11T09:38:44Z
dc.date.available
2023-04-07T06:07:26Z
dc.date.available
2023-04-11T09:38:44Z
dc.date.issued
2022-10
dc.identifier.isbn
978-1-4503-9203-7
en_US
dc.identifier.other
10.1145/3503161.3547919
en_US
dc.identifier.uri
http://hdl.handle.net/20.500.11850/606986
dc.description.abstract
This paper solves the problem of learning image semantic segmentation using image-level supervision. The task is promising in terms of reducing annotation efforts, yet extremely challenging due to the difficulty to directly associate high-level concepts with low-level appearance. While current efforts handle each concept independently, we take a broader perspective to harvest implicit, holistic structures of semantic concepts, which express valuable prior knowledge for accurate concept grounding. This raises multi-granular semantic mining, a new formalism allowing flexible specification of complex relations in the label space. In particular, we propose a heterogeneous graph neural network (Hgnn) to model the heterogeneity of multi-granular semantics within a set of input images. The Hgnn consists of two types of sub-graphs: 1) an external graph characterizes the relations across different images to mine inter-image contexts; and for each image, 2) an internal graph is constructed to mine inter-class semantic dependencies within each individual image. Through heterogeneous graph learning, our Hgnn is able to land a comprehensive understanding of object patterns, leading to more accurate semantic concept grounding. Extensive experimental results show that Hgnn outperforms the current state-of-the-art approaches on the popular PASCAL VOC 2012 and COCO 2014 benchmarks. Our code is available at: https://github.com/maeve07/HGNN.git.
en_US
dc.language.iso
en
en_US
dc.publisher
Association for Computing Machinery
en_US
dc.subject
weakly supervised semantic segmentation
en_US
dc.subject
graph neural networks
en_US
dc.title
Multi-Granular Semantic Mining for Weakly Supervised Semantic Segmentation
en_US
dc.type
Conference Paper
dc.date.published
2022-10-10
ethz.book.title
Proceedings of the 30th ACM International Conference on Multimedia
en_US
ethz.pages.start
6019
en_US
ethz.pages.end
6028
en_US
ethz.event
30th ACM International Conference on Multimedia (MM 2022)
en_US
ethz.event.location
Lisbon, Portugal
en_US
ethz.event.date
October 10-14, 2022
en_US
ethz.identifier.scopus
ethz.publication.place
New York, NY
en_US
ethz.publication.status
published
en_US
ethz.date.deposited
2023-04-07T06:07:26Z
ethz.source
SCOPUS
ethz.eth
yes
en_US
ethz.availability
Metadata only
en_US
ethz.rosetta.installDate
2023-04-11T09:38:45Z
ethz.rosetta.lastUpdated
2023-04-11T09:38:45Z
ethz.rosetta.versionExported
true
ethz.COinS
ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.atitle=Multi-Granular%20Semantic%20Mining%20for%20Weakly%20Supervised%20Semantic%20Segmentation&rft.date=2022-10&rft.spage=6019&rft.epage=6028&rft.au=Zhang,%20Meijie&Li,%20Jianwu&Zhou,%20Tianfei&rft.isbn=978-1-4503-9203-7&rft.genre=proceeding&rft_id=info:doi/10.1145/3503161.3547919&rft.btitle=Proceedings%20of%20the%2030th%20ACM%20International%20Conference%20on%20Multimedia
 Search print copy at ETH Library

Files in this item

FilesSizeFormatOpen in viewer

There are no files associated with this item.

Publication type

Show simple item record