Photobox
An AI powered personalized photo management app
The problem
In recent years, all manufacturers have regarded the camera function as the main function of mobile phones. Through the improvement of both software and hardware, the quality of photography has also been continuously improved. It can be said that mobile phones with cameras have been the most commonly used to record life.
However, in a previous study, the authors claim that for many people, the ability to collect more digital information is not matched by the ability to organize information. Faced with a large number of photos, how to properly manage and categorize photos has become a difficult problem for everyone.
My role
This is a school project, I cooperate with two classmates, from whom, one is focus on research and another is focus on algorithm. I completed all the design work including user interface design and user experience design. In the research part, I complete the whole design work of the questionnaire and sent out 98 questionnaires to collect user data for research analysis.
Design Concept
Our research is to design a mobile phone application, Photobox, that helps users manage their photos on mobile devices.
Generative Research - Pain Points
To better understand users’ needs, we structured a short survey with 15 questions and spread it through Amazon Mechanic Turk. We got 316 responses in total with about an even distribution of males and females.
Insights
Most users would like to improve the manual classification/ tag and photo search system in the current photo management apps, because they are currently not able to annotate or manage tags for photos.
The current photo search system is too simple and not useful since sometimes user should scroll a lot to find the exact picture.
Many respondents would like to have some smart tagging features for events or activities and remove similar photos on their phone.
Generative Research - Features
To make sure that our outcome would be valuable, we decided to focus on just one feature that would be the biggest differentiator for our app and take it further with more applicability. The feature we considered to focus on is similar photo detection, which many current apps started to consider but haven’t provided a reliable solution. To see whether this feature is worth focusing on, we conducted another online survey and structured several interviews to understand whether our users would value this feature.
Result
65% of respondents say that they would take multiple photos with the same scene.
73% find that it is difficult to select the photo they would like to keep or share.
Almost all respondents (94%) think they’d like to see some smart features to help them with detecting similar photos and recommending which one could be the best one to keep.
Design concept & Persona
Based on the result, we planed to build our design with three main features, intelligent tag recommendation, manual correction for the algorithm, and similar photo detection, that would address the problems that users are facing. And I made the persona as bellow.
Storyboard
In order to express our ideas well, I created a storyboard.
1. Taking pictures. 2. It’s difficult to find the best photo in many similar photos. 3. The “Photobox” helps you to detect similar photos and give a recommendation of the best photo. 4. Upload the photo recommended by AI to social media and get some good feedback.
