1 / 1

Photo-quality Assessment and Enhancement based on Visual Aesthetics

Photo-quality Assessment and Enhancement based on Visual Aesthetics. Subhabrata Bhattacharya, (U of Central Florida), Rahul Sukthankar (Intel Labs, Pittsburgh), and Mubarak Shah, (U of Central Florida). Motivation. Approach Overview. Modeling Aesthetics.

nan
Download Presentation

Photo-quality Assessment and Enhancement based on Visual Aesthetics

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Photo-quality Assessment and Enhancement based on Visual Aesthetics Subhabrata Bhattacharya, (U of Central Florida), RahulSukthankar(Intel Labs, Pittsburgh), and Mubarak Shah, (U of Central Florida) Motivation Approach Overview Modeling Aesthetics • Plethora of Digital photo in Personal collections • Not all “good” - Which ones to keep? • Can we make the ones better that are not so “good”? • Goal: Assess photographic quality of an image and suggest alterations to user to make it better • Dataset ~ 384 single subject photos, 248 sea/landscapes • 15 independent human raters • Uni-modal distribution of ranks – averaged to obtain • Independent of equipment capability (depth-of-field etc.) • Addresses Broader class of images • Attempts to solve both assessment and enhancement problems together Aesthetic Features Enhancement algorithms Find optimal location ensuring tree stays in ground When horizon is available, scale object to correct perspective • Scaling factor Optimal Location Find vertical extent to be modified to make the visual weights close to the golden ratio; inpaint the newly added region Map aesthetic features to appeal factor using SVR Results • 87% of the single subject compositions are assessed accurately • 91% of the landscape/seascape images are assessed accurately

More Related