Zero-shot object counting with good exemplars
Zero-shot object counting (ZOC) aims to enumerate objects in images using only the names of object classes during testing, without the need for manual annotations. However, a critical challenge in current ZOC methods lies in their inability to identify high-quality exemplars effectively. This defici...
Saved in:
Main Authors: | , , , , , , |
---|---|
Format: | text |
Language: | English |
Published: |
Institutional Knowledge at Singapore Management University
2024
|
Subjects: | |
Online Access: | https://ink.library.smu.edu.sg/sis_research/9768 https://ink.library.smu.edu.sg/context/sis_research/article/10768/viewcontent/2407.04948v2.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Singapore Management University |
Language: | English |
id |
sg-smu-ink.sis_research-10768 |
---|---|
record_format |
dspace |
spelling |
sg-smu-ink.sis_research-107682024-12-16T02:39:06Z Zero-shot object counting with good exemplars ZHU, Huilin YUAN, Jingling YANG, Zhengwei GUO, Yu WANG, Zheng ZHONG, Xian HE, Shengfeng Zero-shot object counting (ZOC) aims to enumerate objects in images using only the names of object classes during testing, without the need for manual annotations. However, a critical challenge in current ZOC methods lies in their inability to identify high-quality exemplars effectively. This deficiency hampers scalability across diverse classes and undermines the development of strong visual associations between the identified classes and image content. To this end, we propose the Visual Association-based Zero-shot Object Counting (VA-Count) framework. VACount consists of an Exemplar Enhancement Module (EEM) and a Noise Suppression Module (NSM) that synergistically refine the process of class exemplar identification while minimizing the consequences of incorrect object identification. The EEM utilizes advanced vision-language pretaining models to discover potential exemplars, ensuring the framework’s adaptability to various classes. Meanwhile, the NSM employs contrastive learning to differentiate between optimal and suboptimal exemplar pairs, reducing the negative effects of erroneous exemplars. VA-Count demonstrates its effectiveness and scalability in zero-shot contexts with superior performance on two object counting datasets. 2024-10-01T07:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/9768 info:doi/10.1007/978-3-031-72652-1_22 https://ink.library.smu.edu.sg/context/sis_research/article/10768/viewcontent/2407.04948v2.pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University Zero-shot object counting Object classification Object counting framework Class exemplar identification Artificial Intelligence and Robotics Graphics and Human Computer Interfaces |
institution |
Singapore Management University |
building |
SMU Libraries |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
SMU Libraries |
collection |
InK@SMU |
language |
English |
topic |
Zero-shot object counting Object classification Object counting framework Class exemplar identification Artificial Intelligence and Robotics Graphics and Human Computer Interfaces |
spellingShingle |
Zero-shot object counting Object classification Object counting framework Class exemplar identification Artificial Intelligence and Robotics Graphics and Human Computer Interfaces ZHU, Huilin YUAN, Jingling YANG, Zhengwei GUO, Yu WANG, Zheng ZHONG, Xian HE, Shengfeng Zero-shot object counting with good exemplars |
description |
Zero-shot object counting (ZOC) aims to enumerate objects in images using only the names of object classes during testing, without the need for manual annotations. However, a critical challenge in current ZOC methods lies in their inability to identify high-quality exemplars effectively. This deficiency hampers scalability across diverse classes and undermines the development of strong visual associations between the identified classes and image content. To this end, we propose the Visual Association-based Zero-shot Object Counting (VA-Count) framework. VACount consists of an Exemplar Enhancement Module (EEM) and a Noise Suppression Module (NSM) that synergistically refine the process of class exemplar identification while minimizing the consequences of incorrect object identification. The EEM utilizes advanced vision-language pretaining models to discover potential exemplars, ensuring the framework’s adaptability to various classes. Meanwhile, the NSM employs contrastive learning to differentiate between optimal and suboptimal exemplar pairs, reducing the negative effects of erroneous exemplars. VA-Count demonstrates its effectiveness and scalability in zero-shot contexts with superior performance on two object counting datasets. |
format |
text |
author |
ZHU, Huilin YUAN, Jingling YANG, Zhengwei GUO, Yu WANG, Zheng ZHONG, Xian HE, Shengfeng |
author_facet |
ZHU, Huilin YUAN, Jingling YANG, Zhengwei GUO, Yu WANG, Zheng ZHONG, Xian HE, Shengfeng |
author_sort |
ZHU, Huilin |
title |
Zero-shot object counting with good exemplars |
title_short |
Zero-shot object counting with good exemplars |
title_full |
Zero-shot object counting with good exemplars |
title_fullStr |
Zero-shot object counting with good exemplars |
title_full_unstemmed |
Zero-shot object counting with good exemplars |
title_sort |
zero-shot object counting with good exemplars |
publisher |
Institutional Knowledge at Singapore Management University |
publishDate |
2024 |
url |
https://ink.library.smu.edu.sg/sis_research/9768 https://ink.library.smu.edu.sg/context/sis_research/article/10768/viewcontent/2407.04948v2.pdf |
_version_ |
1819113133226065920 |