390 likes | 540 Views
Photo-Quality Enhancement based on Visual Aesthetics. S. Bhattacharya*, R. Sukthankar **, M.Shah * *University of Central Florida, **Intel labs & CMU. Motivation. Outline. Related work Quality enhancement framework Visual aesthetics Aesthetic appeal assessment
E N D
Photo-Quality Enhancement based on Visual Aesthetics S. Bhattacharya*, R. Sukthankar**, M.Shah* *University of Central Florida, **Intel labs & CMU
Outline • Related work • Quality enhancement framework • Visual aesthetics • Aesthetic appeal assessment • Enhancement through recomposition • Experimental results • Conclusions • Future directions
Related Work • Low-level (dehazing etc.) K. He, J.Sun X. Tang, “Single Image Haze Removal using Dark Channel Prior”, CVPR 09 • Domain-specific (face beautification etc.) T.Leyvand,et al. , “Data-Driven Enhancement of Facial Attractiveness”, ACM SIGGRAPH 08
Overview Assessment Engine Aesthetic Features Appeal Prediction Image Semantics, Aesthetic Features Aesthetic Model Input Image Enhanced Image Recomposition Enhancement Engine
Visual Aesthetics: Rule of Thirds Motivated by Renaissance Paintings… http://hoocher.com/Joseph_William_Turner/Joseph_William_Turner.htm Rule of thirds: Subject of interest is aligned to one of the stress points Professional photographs also abide this: http://howtophotography.org/wp-content/uploads/2010/06/rule-of-thirds-photo2.jpg
Visual Aesthetics: Golden Ratio ~1.618k Sky Sky Land Sea k http://hoocher.com/Joseph_William_Turner/Joseph_William_Turner.htm Divine proportion: Horizon divides sky and sea/land according to golden ratio. An example professional photographic composition: http://www.dptips-central.com/rules-of-composition.html
Modeling Aesthetics: Dataset Single subject Compositions (384) Landscapes/Seascapes (248) http://www.flickr.com
Modeling Aesthetics: User study … Rank Assignment between 1-5 1 2 15 14 Single subjects Landscapes/Seascapes Ground Truth Appeal Factors http://www.flickr.com
Modeling Aesthetics: User study 1.76 4.23 Best rated images Poorly rated images
Modeling Aesthetics: User study Poor compositions Good Compositions User Agreements Appeal Factor Intervals
Modeling Aesthetics: Features (a) Relative Foreground Location (Rule of Thirds) Visual Attention Center Stress Point
Modeling Aesthetics: Features (b) Visual weight deviation from Golden Ratio (Divine Proportion) Yk Yg
Experiments (Assessment) • Learn Support Vector Regression models • Prediction accuracy: • Single subject compositions ~ 87% • Landscapes/Seascapes ~ 91% Smooth mapping between Appeal factor and Aesthetic Features • Relative Foreground Location • Visual Weight Deviation
Why Cropping does not work? Optimal Crop
Recomposition: Algorithm I Input Image Optimal Object Placement Spatial Recomposition In-painting Semantic Segmentation Single Subject? Labeled Elements
Semantic Segmentation Input Image Segmented Foreground Sky Geometric Context Classifier* Post Processing Horizon Support *D. Hoiem, A.A. Efros, and M. Hebert, "Geometric Context from a Single Image", ICCV 2005
Optimal Object Placement Find x that Maximizes Appeal Support Neighborhood Labeled Image s.t. neighbors stay “like neighbors” + Intensity Term Gradient Term
Optimization (Example) X PAF = 3.22 Original Image Semantic constraint prevents this Optimal Solution PAF = 3.31 PAF = 4.53 PAF = 3.68
Perspective Scaling Scaled Foreground Visual Attention Center Scaling Factor Optimal location Vanishing Point
Inpainting Foreground Hole Inpaint Hole • Yunjun Zhang. Jiangjian Xiao. Mubarak Shah, “Region Completion in a Single Image”, EUROGRAPHICS 04
Recomposition: Algorithm 2 Input Image Semantic Segmentation Land/Sea scape? Optimally Crop/Expand Labeled Elements Visual Weight Balancing
Balancing Visual Weights • Ratio of Current extents Yk+h Yk • h= vertical extent of the balanced image Yg Yg • Solve for h (sign of h determines crop/expansion)
Experimental Results • Single Subject Composition • Horse is moved to a more visually pleasing location • Scaled appropriately • Appeal increases by 64% • Before Recomposition • After Recomposition
Results PAF = 4.29 PAF = 2.45 • Before • After PAF = 3.98 PAF = 4.46
Results PAF = 4.02 PAF = 4.34 • Before • After PAF = 3.13 PAF = 4.19
Results PAF = 3.77 PAF = 4.25 • Before • After PAF = 3.92 PAF = 4.11
Results PAF = 2.71 PAF = 3.26 • Before • After PAF = 4.68 PAF = 4.06
Results • Optimally cropped support region to increase weights for sky • Appeal factor increased by 51% • Visual weight balancing • Before Recomposition • After Recomposition
Balancing Visual weights PAF = 4.02 PAF = 3.83 • Before • After PAF = 4.38 PAF = 3.92
Balancing Visual weights PAF = 4.71 PAF = 4.02 • Before • After PAF = 4.49 PAF = 4.17
Not Perfect • Algorithm says nice, humans: otherwise Fa = 2.41 (Ground Truth) PAF = 2.34 Fa = 2.54 (Ground Truth) PAF = 3.63 • Before • After
Summary: Optimal Placement • Increased # of Highly rated Images • Decreased # of Poorly rated Images • Before • After
Summary: Visual Weights • Increased # of Highly rated Images • Decreased # of Poorly rated Images • Before • After
Conclusion • Intelligent photo recomposition • Can also be used for aesthetic filtering • Easy to use practical tool
Future Work • Synthesizing ideal image from many photos of the same scene • Recompositionfor videos