Supporting Blind Photography
Chandrika Jayant, Hanjie Ji, Samuel White, Jeffrey P. Bigham · 2011 · ASSETS '11: The Proceedings of the 13th International ACM SIGACCESS Conference on Computers and Accessibility · doi:10.1145/2049536.2049573
Summary
This paper explores how blind and low-vision people use cameras and presents two applications designed to make photography more accessible. The research begins with a large-scale online survey of 118 blind and low-vision people that establishes the empirical foundation for the work. The survey revealed that 71% of respondents had recently used a camera, with 76.3% believing accessible camera use would be valuable. The primary reason for photography was capturing friends, family, trips, and fun (61.9%), followed by text recognition (42.9%). The authors then introduce EasySnap, an iPhone application with three modes: "Freestyle" (standard camera), "People" (uses face detection to provide audio feedback about face location and size), and "Object" (uses SURF feature tracking to help maintain framing of objects). EasySnap also includes an accessible photo album that automatically tags images with GPS location and content recognition via IQEngine, plus exposure and sharpness detection that warns users about blurry or dark photos. Building on EasySnap, the authors developed PortraitFramer, an Android application specifically focused on group portrait photography. PortraitFramer uses face detection to announce the number of faces found, displays high-contrast circles on a black screen where faces are detected, provides vibration when face areas are touched on the touchscreen, uses distinct pitches for each face, and offers optional directional instructions for centering the composition.
Key findings
The survey of 118 blind and low-vision respondents showed that even among totally blind respondents, 72.7% had recently used a camera — a finding the researchers themselves found surprising. In the EasySnap evaluation with 6 participants, 31 unbiased sighted viewers judged 1,116 photo pairs and preferred photos taken with audio feedback 58% of the time versus 29% without feedback (12% neutral), with both Object and People modes achieving approximately 60% success rates. Participants rated EasySnap highly on Likert scales for helpfulness and ease of use. In the first PortraitFramer study with 8 blind and low-vision participants, all successfully centered faces after receiving vibration and overlay cues, with an average time of 13 seconds per successful photo and all participants succeeding within three attempts. In the second PortraitFramer study with 7 participants (no overlap with first study), using the instruction mode all participants took a successfully centered photo within an average of 5 seconds and 3 tries. The "I would use this application" Likert score improved significantly from the first version (average < 4) to the second version (average close to 6 on a 7-point scale, t = 2.15, DF = 13, p < 0.05). Participants were overwhelmingly positive, with comments like "the app makes me feel confident that I didn't chop the heads off" and "Now I can take a picture."
Relevance
This paper makes a compelling case that accessibility extends beyond functional task completion to creative and social activities. The finding that blind people already actively take photographs — for artistic expression, memory preservation, and social sharing — challenges assumptions about which technologies are relevant to blind users. For accessibility practitioners, this research demonstrates the importance of multimodal feedback design: combining audio instructions, haptic vibrations, distinct pitches, and high-contrast visual overlays to serve users across a spectrum of vision levels. The study also surfaces important social dimensions of accessible technology, including participants' concerns about public perception of blind people using cameras, the desire for autonomy versus detailed guidance (some wanted minimal cues for speed, others wanted explicit instructions), and the importance of customization options. The work anticipates the broader trend of making smartphone cameras accessible and has clear connections to modern applications like Be My Eyes and seeing AI tools that use cameras as input for visual information access.
Tags: blind users · low vision · photography · computer vision · mobile accessibility · audio feedback · haptic feedback · face detection · assistive technology