Learning to compose with professional photographs on the web

Yi Ling Chen, Jan Klopp, Min Sun, Shao Yi Chien, Kwan-Liu Ma

Research output: Chapter in Book/Report/Conference proceedingConference contribution

12 Citations (Scopus)

Abstract

Photo composition is an important factor affecting the aesthetics in photography. However, it is a highly challenging task to model the aesthetic properties of good compositions due to the lack of globally applicable rules to the wide variety of photographic styles. Inspired by the thinking process of photo taking, we formulate the photo composition problem as a view finding process which successively examines pairs of views and determines their aesthetic preferences. We further exploit the rich professional photographs on the web to mine unlimited high-quality ranking samples and demonstrate that an aesthetics-aware deep ranking network can be trained without explicitly modeling any photographic rules. The resulting model is simple and effective in terms of its architectural design and data sampling method. It is also generic since it naturally learns any photographic rules implicitly encoded in professional photographs. The experiments show that the proposed view finding network achieves state-of-the-art performance with sliding window search strategy on two image cropping datasets.

Original languageEnglish (US)
Title of host publicationMM 2017 - Proceedings of the 2017 ACM Multimedia Conference
PublisherAssociation for Computing Machinery, Inc
Pages37-45
Number of pages9
ISBN (Electronic)9781450349062
DOIs
StatePublished - Oct 23 2017
Event25th ACM International Conference on Multimedia, MM 2017 - Mountain View, United States
Duration: Oct 23 2017Oct 27 2017

Other

Other25th ACM International Conference on Multimedia, MM 2017
CountryUnited States
CityMountain View
Period10/23/1710/27/17

Fingerprint

Photocomposition
Architectural design
Photography
Sampling
Chemical analysis
Experiments

Keywords

  • Aesthetics modeling
  • Deep ranking network
  • Photo composition
  • View recommendation

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Media Technology
  • Computer Vision and Pattern Recognition
  • Software

Cite this

Chen, Y. L., Klopp, J., Sun, M., Chien, S. Y., & Ma, K-L. (2017). Learning to compose with professional photographs on the web. In MM 2017 - Proceedings of the 2017 ACM Multimedia Conference (pp. 37-45). Association for Computing Machinery, Inc. https://doi.org/10.1145/3123266.3123274

Learning to compose with professional photographs on the web. / Chen, Yi Ling; Klopp, Jan; Sun, Min; Chien, Shao Yi; Ma, Kwan-Liu.

MM 2017 - Proceedings of the 2017 ACM Multimedia Conference. Association for Computing Machinery, Inc, 2017. p. 37-45.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Chen, YL, Klopp, J, Sun, M, Chien, SY & Ma, K-L 2017, Learning to compose with professional photographs on the web. in MM 2017 - Proceedings of the 2017 ACM Multimedia Conference. Association for Computing Machinery, Inc, pp. 37-45, 25th ACM International Conference on Multimedia, MM 2017, Mountain View, United States, 10/23/17. https://doi.org/10.1145/3123266.3123274
Chen YL, Klopp J, Sun M, Chien SY, Ma K-L. Learning to compose with professional photographs on the web. In MM 2017 - Proceedings of the 2017 ACM Multimedia Conference. Association for Computing Machinery, Inc. 2017. p. 37-45 https://doi.org/10.1145/3123266.3123274
Chen, Yi Ling ; Klopp, Jan ; Sun, Min ; Chien, Shao Yi ; Ma, Kwan-Liu. / Learning to compose with professional photographs on the web. MM 2017 - Proceedings of the 2017 ACM Multimedia Conference. Association for Computing Machinery, Inc, 2017. pp. 37-45
@inproceedings{6b966a66681149b690e494304f5d6615,
title = "Learning to compose with professional photographs on the web",
abstract = "Photo composition is an important factor affecting the aesthetics in photography. However, it is a highly challenging task to model the aesthetic properties of good compositions due to the lack of globally applicable rules to the wide variety of photographic styles. Inspired by the thinking process of photo taking, we formulate the photo composition problem as a view finding process which successively examines pairs of views and determines their aesthetic preferences. We further exploit the rich professional photographs on the web to mine unlimited high-quality ranking samples and demonstrate that an aesthetics-aware deep ranking network can be trained without explicitly modeling any photographic rules. The resulting model is simple and effective in terms of its architectural design and data sampling method. It is also generic since it naturally learns any photographic rules implicitly encoded in professional photographs. The experiments show that the proposed view finding network achieves state-of-the-art performance with sliding window search strategy on two image cropping datasets.",
keywords = "Aesthetics modeling, Deep ranking network, Photo composition, View recommendation",
author = "Chen, {Yi Ling} and Jan Klopp and Min Sun and Chien, {Shao Yi} and Kwan-Liu Ma",
year = "2017",
month = "10",
day = "23",
doi = "10.1145/3123266.3123274",
language = "English (US)",
pages = "37--45",
booktitle = "MM 2017 - Proceedings of the 2017 ACM Multimedia Conference",
publisher = "Association for Computing Machinery, Inc",

}

TY - GEN

T1 - Learning to compose with professional photographs on the web

AU - Chen, Yi Ling

AU - Klopp, Jan

AU - Sun, Min

AU - Chien, Shao Yi

AU - Ma, Kwan-Liu

PY - 2017/10/23

Y1 - 2017/10/23

N2 - Photo composition is an important factor affecting the aesthetics in photography. However, it is a highly challenging task to model the aesthetic properties of good compositions due to the lack of globally applicable rules to the wide variety of photographic styles. Inspired by the thinking process of photo taking, we formulate the photo composition problem as a view finding process which successively examines pairs of views and determines their aesthetic preferences. We further exploit the rich professional photographs on the web to mine unlimited high-quality ranking samples and demonstrate that an aesthetics-aware deep ranking network can be trained without explicitly modeling any photographic rules. The resulting model is simple and effective in terms of its architectural design and data sampling method. It is also generic since it naturally learns any photographic rules implicitly encoded in professional photographs. The experiments show that the proposed view finding network achieves state-of-the-art performance with sliding window search strategy on two image cropping datasets.

AB - Photo composition is an important factor affecting the aesthetics in photography. However, it is a highly challenging task to model the aesthetic properties of good compositions due to the lack of globally applicable rules to the wide variety of photographic styles. Inspired by the thinking process of photo taking, we formulate the photo composition problem as a view finding process which successively examines pairs of views and determines their aesthetic preferences. We further exploit the rich professional photographs on the web to mine unlimited high-quality ranking samples and demonstrate that an aesthetics-aware deep ranking network can be trained without explicitly modeling any photographic rules. The resulting model is simple and effective in terms of its architectural design and data sampling method. It is also generic since it naturally learns any photographic rules implicitly encoded in professional photographs. The experiments show that the proposed view finding network achieves state-of-the-art performance with sliding window search strategy on two image cropping datasets.

KW - Aesthetics modeling

KW - Deep ranking network

KW - Photo composition

KW - View recommendation

UR - http://www.scopus.com/inward/record.url?scp=85035190725&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85035190725&partnerID=8YFLogxK

U2 - 10.1145/3123266.3123274

DO - 10.1145/3123266.3123274

M3 - Conference contribution

AN - SCOPUS:85035190725

SP - 37

EP - 45

BT - MM 2017 - Proceedings of the 2017 ACM Multimedia Conference

PB - Association for Computing Machinery, Inc

ER -