<html><body><div style="font-family: garamond,new york,times,serif; font-size: 12pt; color: #000000"><div>Ohad Fried will present his FPO on Wednesday, May 17, 2017 at 1:30pm in CS 401.</div><div><br data-mce-bogus="1"></div><div>The members of his committee are: Adam Finkelstein (adviser); readers: Tom Funkhouser and Eli Shechtman (Adobe); examiners: Szymon Rusinkiewicz and David Dobkin.</div><div><br>A copy of his thesis is available in Room 310.</div><div><br>Everyone is invited to attend his talk. The talk title and abstract follow below:<br></div><div><br data-mce-bogus="1"></div><div>PHOTO MANIPULATION, THE EASY WAY</div><div>The typical smartphone user has many thousands of photos in their personal collection.<br>Photo acquisition is effortless, and the next challenge is in devising methods to easily edit<br>such large collections. Specifically, we need manipulation algorithms that are powerful<br>enough for experts, yet simple for novices to master.<br>We identified three key directions to empower novice users with expert-level editing<br>capabilities while maintaining an overall simplicity in the process. Those directions are<br>(1) better selection masks, (2) high-level goal-centric algorithms and (3) domain specific<br>algorithms. In this thesis we give examples from each category.<br>Given a photo, a novice user will typically either not edit it at all, or apply a simple<br>global operation such as exposure correction. In contrast, a professional photo editor<br>might perform local edits, specifying a selection mask to limit the operation to specific<br>photo regions, or combining regions from several photos into a single composition. To<br>ease selection mask creation we present a new patch embedding technique that allows for<br>single-click selection masks.<br>Novices often think in terms of goals (e.g. improve lighting, de-clutter photo) and less<br>in technical terms such as color spaces and image layers. One example of a high-level<br>goal is the removal of distracting elements from photos. The task is motivated by the<br>way professional photographers operate. They carefully frame the scene and might move<br>objects around in order to stage the perfect photo. We define “photo distractors” as the<br>elements that, if removed, would improve the photo. Using a simple slider interaction we<br>allow users to automatically remove such distractors from photos.<br>It is at times useful to tailor solutions to specific photo types. As an example we show<br>that, specifically for human heads, simple controls can induce sophisticated edits. Given a<br>single portrait photo as input, we can change the pose of the head and the camera distance.<br>This allows users to correct the “selfie effect”, i.e. big noses and small ears or to transform<br>distant photos into selfies.<br></div><div>We conclude by discussing how each of these directions can be further explored to<br>enable better image editing tools for novices.</div></div></body></html>