“It’s Kind of Context Dependent”: Understanding Blind and Low Vision People’s Video Accessibility Preferences Across Viewing Scenarios

Title of the Paper

“It’s Kind of Context Dependent”: Understanding Blind and Low Vision People’s Video Accessibility Preferences Across Viewing Scenarios

Paper Information

• Subject Area: Accessibility Technology and Human-Computer Interaction
• Keywords: Video Accessibility, Audio Description, Blindness, Low Vision, Context, Audio Cues, Haptic Feedback, Context Awareness, Context-Based Design

Research Background and Problems

• What problems or challenges did the authors identify?
  Current video accessibility solutions primarily focus on generic audio descriptions (AD) and fail to adequately address the diverse needs of blind and low vision (BLV) users across different viewing scenarios. Meanwhile, the variety of video content (e.g., tutorial videos, historical dramas, short videos) continues to grow, as does the diversity of viewing devices and platforms (e.g., mobile devices, YouTube, Instagram).

• Why is this problem important?
  Video, as a mainstream modern information medium, has become a central platform for learning, entertainment, and social interaction. However, due to the inadequacies of existing technologies and designs in addressing the accessibility needs of BLV users, these users are still unable to fully enjoy the diverse experiences that video playback offers.

• Research Motivation and Related Work
  Previous research has primarily focused on general guidelines for video descriptions and content generation but lacks systematic exploration of personalized solutions tailored to specific contexts. Moreover, the increasingly diverse scenarios of video consumption highlight the growing need for context-specific accessibility design.

Solutions

• What methods or solutions did the authors propose?
  Through a survey (101 participants) and 15 semi-structured interviews, the study investigated BLV users’ video accessibility preferences across various contexts. The authors developed a six-dimensional design space to guide future innovations in video accessibility, covering levels of detail, timing adjustments, enhancement methods, presentation modalities, content synchronization, and stylistic tone.

• What are the innovative aspects of this solution?

  1. Moving beyond a "one-size-fits-all" approach by linking video accessibility design to specific contexts.
  2. Including multimodal outputs (e.g., audio cues, tactile graphics, 3D models) as options for video accessibility design.
  3. Proposing a six-dimensional design space for video accessibility, encompassing factors from content granularity to synchronization modes.

• What are the implementation steps? What key technologies were used?

  1. Preliminary survey: Investigated BLV users’ primary video-watching behaviors and barriers through a questionnaire.
  2. Interview study: Conducted in-depth exploration of different video contexts (e.g., tutorial videos, entertainment videos, social media short videos).
  3. Data analysis: Used thematic analysis to summarize common needs and preferences of participants.
  4. Categorization and synthesis: Organized accessibility improvement suggestions from participants into a six-dimensional design framework.

Research Findings

• What specific outcomes were achieved?

  1. Identified differences in BLV users’ needs for video description content across various contexts:
     • Tutorial videos: Focused on operational details and device instructions.
     • Educational videos: Desired more details on visual aids (e.g., infographics), scene settings, and themes.
     • Social short videos: Emphasized characters, actions, and background context.
  2. Proposed a six-dimensional design space for video accessibility:
     • Levels of detail (detailed vs. concise).
     • Timing adjustments (no change vs. added extended content).
     • Degree of enhancement (no enhancement vs. multiple supplementary elements).
     • Presentation modalities (e.g., audio description, visual enhancement, haptic feedback).
     • Content synchronization (before, during, or after the video).
     • Stylistic tone (neutral, subjective, humorous, etc.).
  3. Quantified user acceptance of emerging output modalities (e.g., 3D models, tactile graphics) through interview examples.

• What advantages does it offer compared to existing solutions?

  1. Greater emphasis on contextual relevance, providing personalized accessibility solutions.
  2. Encourages design expansion from purely auditory descriptions to multimodal outputs, enhancing user immersion.
  3. Balances technological advancements (e.g., generative AI applications) with human-centered needs, offering directions for future innovation.

• What were the experimental or evaluation results?
  The six-dimensional design space was well-received by participants, with most expressing approval of context-driven accessibility improvements. Additionally, specific resources prepared for particular video types, tactile enhancements, and audio cues significantly improved users’ comprehension of content.

• Limitations and Future Directions

  1. The interview participants were predominantly blind users, with limited representation of low vision perspectives.
  2. Due to dataset and technological constraints, the study’s coverage of short videos was not comprehensive.
  3. Future research should explore the potential of generative AI in video accessibility design while addressing ethical concerns such as bias and privacy risks.