SelPh: Progressive Learning and Support of Manual Photo Color Enhancement
summary
SelPh is a working prototype that progressively learns a user’s manual color edits and applies five support functions including VisOpt Slider visualization, personalized auto-enhancement, interactive slider optimization, and a confidence value that adapts assistance. It learns a personalized distance metric from the editing history, updates after each photo, and was tested with eight photographers who shot 100 photos then edited 50 with a baseline and 50 with SelPh; results show marginally faster completion and significantly fewer slider moves, and strong user approval.
AI-written summary · about
chapters (9)
- 0:00Intro
- 0:56Problem Statement
- 2:24Workflow
- 4:04Related Works
- 6:31System Details
- 8:18Algorithms
- 10:22Evaluation
- 11:38Results
- 14:18Summary
related talks
08mDynamic Guidance for Decluttering Photographic Compositions
07m#ProjectScribbler
25mPersonalizing Your Images with AI-Powered Features
20mGoogle Photos Magic Editor: GenAI Under the Hood of a Billion-User App
37mMedia AI Photo Studio
11mHuman-centered technology and the role of Responsible AI
37mEnhancing Media Experiences in the Age of AI: A Practical Guide
25mRemoving Soft Shadows with Hard Data