Junfeng Ye

dblp:123/2924 · DBLP profile ↗
← Back
3ranked-venue papers
2as first author
1since 2021 · last 2027
—ORCID · conflict

Domains — the database's venue-derived domains; a paper can count in several

Graphics, computer vision, multimedia, augmented reality and games · 2 · 2 first-authorArtificial intelligence and machine learning · 1 · 1 since 2021

Expertise — from the expertise taxonomy: the topics of the expert's papers under the CCF categories. A weight counts papers with recency: 1 for a paper about the topic, 0.3 when the topic is its context, halved every five years.

Computer graphics and multimedia
2 papers
Multimedia analysis and retrieval · 100%
Databases, data mining, and information retrieval
2 papers
Information retrieval · 100%

Topics — the 3 heaviest of 3, each with the papers that count most for it

TopicWeightPapersLastEvidence papers
Multimedia analysis and retrieval › image retrieval › instance-level image retrieval
landmark search
0.322012
Searching for diversified landmarks by photo · ACM Multimedia 2012
DLMSearch: diversified landmark search by photo · ACM Multimedia 2012
Information retrieval
diversified retrieval
0.222012
DLMSearch: diversified landmark search by photo · ACM Multimedia 2012
Searching for diversified landmarks by photo · ACM Multimedia 2012
Multimedia analysis and retrieval
visual summarization
0.012012
DLMSearch: diversified landmark search by photo · ACM Multimedia 2012
YearPublicationVenuePosition
2027 OWseg-WE: An effective method for open-world semantic segmentation of ginkgo biloba pests and diseases
Yuebo Zhou, Junfeng Ye, Xuemei Xie, Yurong Sun
Expert Syst. Appl.2
2012 DLMSearch: diversified landmark search by photo
abstract
This paper focuses on the problem of searching for diversified landmarks with photos. More particularly, we propose a system called DLMSearch which handles image query, searches for diversified landmarks and provides representative visual summaries. DLMSearch allows a user to upload a query photo and searches for landmarks with high relevance and diversity in real time. Then DLMSearch presents a delicate photo summary for each returned landmark, considering both visual representativeness and diversity. Quantative evaluations on a web-scale landmark photo collection demonstrate the effectiveness of the DLMSearch system. Experimental results verify the merits of the proposed system.
Junfeng Ye, Jia Chen 0001, Zejia Chen, Yihe Zhu, Shenghua Bao, Zhong Su, Yong Yu 0001
ACM Multimedia1
2012 Searching for diversified landmarks by photo
abstract
This demo focuses on the problem of searching for diversified landmarks with photos as input. More particularly, we propose a system called DLMSearch that allows a user to upload a photo as a query and searches for a diverse set of relevant landmarks in real time. It also presents a photo summary for each retrieved landmark, considering both visual representativeness and diversity. Our online demo is available at http://lm.apexlab.org/landmark/demo.
Junfeng Ye, Jia Chen 0001, Zejia Chen, Yihe Zhu, Shenghua Bao, Zhong Su, Yong Yu 0001
ACM Multimedia1