Peter O'Donovan
ABSTRACT
This paper investigates individual variation in aesthetic preferences, and learns models for predicting the preferences of individual users. Preferences for color aesthetics are learned using a collaborative filtering approach on a large dataset of rated color themes/palettes. To make predictions, matrix factorization is used to estimate latent vectors for users and color themes. We also propose two extensions to the probabilistic matrix factorization framework. We first describe a feature-based model using learned transformations from feature vectors to a latent space, then extend this model to non-linear transformations using a neural network. These extensions allow our model to predict preferences for color themes not present in the training set. We find that our approach for modelling user preferences outperforms an average aesthetic model which ignores personal variation. We also use the model for measuring theme similarity and visualizing the space of color themes.
- Adams, R. P., Dahl, G. E., and Murray, I. 2010. Incorporating Side Information Into Probabilistic Matrix Factorization Using Gaussian Processes. In Proc. UAI.Google Scholar
- Basu, C., Hirsh, H., and Cohen, W. 1998. Recommendation as Classification. In Proc. AAAI. Google ScholarDigital Library
- Bychkovsky, V., Paris, S., Chan, E., and Durand, F. 2011. Learning Photographic Global Tonal Adjustment with a Database of Input / Output Image Pairs. In Proc. CVPR. Google ScholarDigital Library
- Chen, T., Zheng, Z., Lu, Q., Z. W., and Yu, Y. 2011. Feature-based Matrix Factorization. Tech. rep., Apex Data and Knowledge Management Lab, Shanghai.Google Scholar
- Datta, R., Joshi, D., Li, J., and Wang, J. Z. 2006. Studying Aaesthetics in Photographic Images Using a Computational Approach. In Proc. ECCV, 7--13. Google ScholarDigital Library
- Granger, G. W. 1952. Objectivity of Color Preferences. Nature 170, 4332.Google ScholarCross Ref
- Kapoor, A., Caicedo, J., Lischinski, D., and Kang, S. 2013. Collaborative Personalization of Image Enhancement. IJCV. Google ScholarDigital Library
- Koren, Y., Bell, R., and Volinsky, C. 2009. Matrix Factorization Techniques for Recommender Systems. IEEE Computer. Google ScholarDigital Library
- Landauer, T. K., and Dumais, S. T. 1997. A Solution to Plato's Problem: The Latent Semantic Analysis Theory of Acquisition, Induction, and Representation of Knowledge. Psychological Review.Google Scholar
- Li, C., and Chen, T. 2009. Aesthetic Visual Quality Assessment of Paintings. Sel. Topics in Signal Processing 3, 2.Google Scholar
- Lin, S., and Hanrahan, P. 2013. Modeling How People Extract Color Themes from Images. In Proc. CHI. Google ScholarDigital Library
- Lin, S., Ritchie, D., Fisher, M., and Hanrahan, P. 2013. Probabilistic Color-by-Numbers: Suggesting Pattern Colorizations Using Factor Graphs.Google Scholar
- Marchesotti, L., Perronnin, F., Larlus, D., and Csurka, G. 2011. Assessing the Aesthetic Quality of Photographs Using Generic Image Descriptors. In Proc. ICCV. Google ScholarDigital Library
- Martindale, C., Moore, K., and Borkum, J. 1990. Aesthetic Preference. The American Journal of Psychology.Google Scholar
- Moorthy, A. K., Obrador, P., and Oliver, N. 2010. Towards Computational Models of Visual Aesthetic Appeal of Consumer Videos. In Proc. ECCV. Google ScholarDigital Library
- O'Donovan, P., Agarwala, A., and Hertzmann, A. 2011. Color Compatibility from Large Datasets. ACM Trans. Graphics (Proc. SIGGRAPH). Google ScholarDigital Library
- Ou, L.-C., and Luo, M. R. 2006. A Color Harmony Model for Two-Color Combinations. Col. Res. Appl 31, 5, 191--204.Google ScholarCross Ref
- Palmer, S., and Griscom, W. 2013. Accounting for Taste: Individual Differences in Preference for Harmony. Psychon Bull Rev.Google Scholar
- Prem Melville, R. M., and Nagarajan, R. 2001. Content-boosted Collaborative Filtering. In Proc. SIGIR Workshop on Recommender Systems.Google Scholar
- Reinecke, K., and Gajos, K. Z. 2014. Quantifying Visual Preferences Around the World. In Proc. CHI. Google ScholarDigital Library
- Salakhutdinov, R., and Mnih, A. 2008. Bayesian Probabilistic Matrix Factorization Using Markov Chain Monte Carlo. In Proc. ICML. Google ScholarDigital Library
- Salakhutdinov, R., and Mnih, A. 2008. Probabilistic Matrix Factorization. In Proc. NIPS.Google Scholar
- Schloss, K. B., and Palmer, S. E. 2010. Aesthetics of Color Combinations. In Human Vision and Elec. Imaging, 752719.Google Scholar
- Su, X., and Khoshgoftaar, T. M. 2009. A Survey of Collaborative Filtering Techniques. Adv. in Artif. Intell. Google ScholarDigital Library
- Szabó, F., Bodrogi, P., and Schanda, J. 2010. Experimental Modeling of Colour Harmony. Col. Res. Appl 35, 1, 34--39.Google ScholarCross Ref
- Tibshirani, R. 1996. Regression Shrinkage and Selection Via the Lasso. Royal. Statist. Soc B 58, 1, 267--288.Google ScholarCross Ref
- van der Maaten, L., and Hinton, G. 2008. Visualizing High-Dimensional Data Using t-SNE. Journal of Machine Learning Research 9, 2579--2605.Google Scholar
- Wang, B., Yu, Y., Wong, T.-T., Chen, C., and Xu, Y.-Q. 2010. Data-Driven Image Color Theme Enhancement. ACM Trans. Graphics (Proc. SIGGRAPH Asia). Google ScholarDigital Library
- Wu, O., Hu, W., and Gao, J. 2011. Learning to Predict the Perceived Visual Quality of Photos. In Proc. ICCV, 225--232. Google ScholarDigital Library
- Yu, L.-F., Yeung, S. K., Terzopoulos, D., and Chan, T. F. 2012. DressUp!: Outfit Synthesis Through Automatic Optimization.Google Scholar
Index Terms
- Collaborative filtering of color aesthetics
Recommendations
An empirical investigation of color temperature and gender effects on web aesthetics
Limited research exists on the relevance of hedonic dimensions of human-computer interaction to usability, with only a small set of this research being empirical in nature. Furthermore, previous research has obtained mixed support for gender differences ...
Color spaces and color contrast
With the introduction of low-cost color graphics systems comes a host of problems specifically concerned with the color aspect of the system. This paper discusses two of these problems: the selection and manipulation of colors by (possibly) ...
Color Space Selection for Color Image Enhancement Applications
ICSAP '09: Proceedings of the 2009 International Conference on Signal Acquisition and ProcessingDevice independent, quantitative description of color is a challenging problem. Another problem is that even under equal intensity, some colors are visually brighter than others. Different color representations try to overcome these problems, with ...
Comments