uPhoto

"Can you take a photo of us in front of the flowers?"

This has happened to all of us. You ask someone to take a photo while on vacation and what you get is a head scratcher. Do you tell the kind gentleman that took your photo that he sucks at taking photos or do you wait for him to go away and ask someone else to try again?

The purpose of uPhoto is to eliminate the guesswork and to get that perfect shot the first time. Every time.

Designed for accessibility but usable by everyone

The idea for uPhoto first sparked during a lunch with an old friend who is partially but legally blind. He was asked to take a photo of a couple on vacation and was eager to help them capture the moment. However, without the right tools, he found it challenging to take an effective picture. That conversation ignited the concept for uPhoto.

As a designer, accessibility is more than just a requirement—it’s a passion woven into every part of my process. While many view accessibility (A11y) as a complex and costly consideration aimed at only a small percentage of users, the reality is that accessible designs enhance the experience for everyone.

My design philosophy is rooted in empathy, and uPhoto is a perfect example of how inclusive design can empower all users, regardless of their abilities.

Exploring Concepts

I spent a significant amount of time exploring different ways people ask others to take photos of them. Reflecting on my own travel experiences, I noticed several factors that affect the quality of the picture—ranging from language barriers to the photographer’s skill and even their age.

While some ideas were more accessible than others, we tested them all, and these winning concepts stood out. Each one incorporated a combination of visual, audible, or tactile feedback, ensuring a more inclusive and user-friendly experience

Wireframes

The first concept I explored utilized NFC, or 'Tap to Pair' technology, to quickly establish a secure, closed Wi-Fi connection between two phones for seamless video streaming. This allowed one user to aim the camera while the other directed the shot in real-time through live video and voice, all over a secure connection.

It proved to be an intuitive and efficient way to capture the perfect shot. However, due to hardware limitations on Apple devices at the time, this approach only worked reliably on Android phones.

A later version that worked on all phones utilized a QR code or pairing code to sync the apps and connection.

Throughout the process, an optional audio narration was available to guide the paring process.

Onion Skin Mode

Onion skinning, a method I used quite a bit when rotoscoping for animation, would allow users to take a framing shot and overlay it on the live camera feed, enabling them to match the original composition by adjusting the camera.

This method was the simplest of all the approaches to develop, but testing revealed issues with contrast that made it difficult to distinguish between the overlay and the live feed. To solve this, I introduced a slider that allowed users to adjust the opacity and amount of overlay of the framing shot, maintaining clarity in the live feed while still using the overlay as a guide.

Later concepts harnessed machine vision and AI for scene matching. By analyzing what the AI detected in the environment, the app offered verbal or visual cues to help users align and match the scene more accurately.

Audio cues were also used to help identify when scene matching was correct. A soft beep would get faster or slower depending on which direction the camera was moved. Faster beeps indicated higher matching while a solid tone indicated good framing of the photo.

Sensor Mode

Another method we explored involved using sensor data such as lidar, magnetometer, GPS, and barometer.

In real time, uPhoto would read data from these sensors to determine the phone’s position in the physical world.

Building on that data, I developed a compass and a level interface to assist the user in properly aligning the phone.

The user would move the phone until a dot was centered on the screen. A photo would then be taken automatically once alignment was detected. Alternatively, the user could take a photo manually by tapping anywhere on the screen or pressing the volume button.

To improve accessibility for my friend, who is blind, we extended the original audio guide by introducing haptic feedback. Like the audio cues, the haptic pulses would increase or decrease in speed depending on the direction the phone was being moved.

While these features were designed to meet the specific needs of my friend, the combination of visual, haptic, and auditory feedback—along with alignment aids—also proved useful for a secondary scenario: helping others understand exactly how someone wants a photo taken, without requiring verbal instructions.

Final Results

We began by using NFC to initiate a secure handshake between two phones, which automatically established a direct Wi-Fi connection. This connection enabled a low-latency VOIP audio link over speakerphone, allowing both users to communicate clearly without the need for additional setup. Once connected, the system launched the camera app on one device and began streaming its live video feed to the second phone. This allowed the photo subject to see exactly what the camera saw, as if they were physically behind it.

With this setup, the person holding the phone simply pointed it while following voice instructions from the subject. The subject could direct framing in real time, giving commands like “a little higher,” “pan left,” or “zoom in slightly.” This created a collaborative, remote-controlled photography experience. The subject not only had a live view but also had full control over the shutter and other camera functions. When they were satisfied with the composition, they could press a button on their own device to take the photo instantly.

The process was fluid and intuitive. After each shot, the subject could review the image, provide feedback, and continue directing adjustments until they achieved exactly the photo they wanted. This approach eliminated guesswork and made it possible for someone with no sight, or someone unfamiliar with the user's preferences, to act as the camera operator under real-time guidance. It turned the experience of taking a photo into a shared, conversational interaction, empowering the subject with full creative control.