TY - JOUR
T1 - Automatic triage for a photo series
AU - Chang, Huiwen
AU - Yu, Fisher
AU - Wang, Jue
AU - Ashley, Douglas
AU - Finkelstein, Adam
N1 - Funding Information:
We thank the anonymous reviewers for helpful suggestions. This work was supported in part by Adobe and the NSF (IIS-1421435). We thank all of the contributors to our photo dataset: Rachel Adler, Doug Ashley, Carolyn Beard, Fei Bo, Guillermo Martinez Cabalga, Ethan Campbell, Anne Caswell-Klein, Adam Cellon, Huiwen Chang, Audrey Chebet, Desai Chen, Stephanie Diu, Rita Fang, Yun Fei, Noah Fishman, Nicolas Freeman, Avigail Gilad, Lisa Gong, Heather Grace, Jesenia Haynes, Sonia Howlett, Joan Hsiao, Emily Hsu, Sohee Hyung, Benjamin Jacobson, Clare Jeong, Jacob Kaplan, Tyler Kaye, Ji-Sung Kim, Tom Kingori, Grace Koh, Jiaming Kong, Taylor Kulp-McDowall, Ashlyn Lackey, Alex Lam, Katherine Lee, Victoria Lin, Kathryn Little, Alex Liu, Cindy Liu, Jennifer Liu, Weber Liu, Annie Lu, Natalie Lu, Angela Mao, Lindsay Martinez, Anna Matlin, Ryan McCaffrey, Mitch Mitchell, John Morone, Shefali Nayak, Anna Pearson, Luke Petruzzi, Selina Pi, Vincent Po, Sahand Keshavarz Rahbar, Andrew Salmons, Leigh Schriever, Paarth Shah, Jessica Shi, Danny Shum, Sarah Spergel, Zachary Stecker, Mark Tengi, Masako Toyoda, Linh Tran, Sarah Tucker, Sadie Van Vranken, Nina Wade, Jue Wang, Kerith Wang, Xue Wang, Ziggie Wang, Samantha Weissman, Jennie Werner, Daniel Wilson, Yihua Xie, Alexander Xu, Dyland Xue, Mandy Yang, Aravind Yeduvaka, Jennifer Yin, Lily Zhang, Maggie Zhang, Mark Zhang, Yinda Zhang, Yuanyuan Zhao, Serena Zheng, Lulu Zhong, Carlos Zhu.
Publisher Copyright:
© 2016 Copyright held by the owner/author(s).
PY - 2016/7/11
Y1 - 2016/7/11
N2 - People often take a series of nearly redundant pictures to capture a moment or scene. However, selecting photos to keep or share from a large collection is a painful chore. To address this problem, we seek a relative quality measure within a series of photos taken of the same scene, which can be used for automatic photo triage. Towards this end, we gather a large dataset comprised of photo series distilled from personal photo albums. The dataset contains 15; 545 unedited photos organized in 5; 953 series. By augmenting this dataset with ground truth human preferences among photos within each series, we establish a benchmark for measuring the effectiveness of algorithmic models of how people select photos. We introduce several new approaches for modeling human preference based on machine learning. We also describe applications for the dataset and predictor, including a smart album viewer, automatic photo enhancement, and providing overviews of video clips.
AB - People often take a series of nearly redundant pictures to capture a moment or scene. However, selecting photos to keep or share from a large collection is a painful chore. To address this problem, we seek a relative quality measure within a series of photos taken of the same scene, which can be used for automatic photo triage. Towards this end, we gather a large dataset comprised of photo series distilled from personal photo albums. The dataset contains 15; 545 unedited photos organized in 5; 953 series. By augmenting this dataset with ground truth human preferences among photos within each series, we establish a benchmark for measuring the effectiveness of algorithmic models of how people select photos. We introduce several new approaches for modeling human preference based on machine learning. We also describe applications for the dataset and predictor, including a smart album viewer, automatic photo enhancement, and providing overviews of video clips.
KW - Benchmark
KW - Photo quality
KW - Photo triage
UR - http://www.scopus.com/inward/record.url?scp=84980004893&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84980004893&partnerID=8YFLogxK
U2 - 10.1145/2897824.2925908
DO - 10.1145/2897824.2925908
M3 - Conference article
AN - SCOPUS:84980004893
SN - 0730-0301
VL - 35
JO - ACM Transactions on Graphics
JF - ACM Transactions on Graphics
IS - 4
M1 - a148
T2 - ACM SIGGRAPH 2016
Y2 - 24 July 2016 through 28 July 2016
ER -