Branch Explorer: Leveraging Branching Narratives to Support Interactive 360° Video Viewing for Bl...
summary
Branch Explorer converts 360° videos into branching narratives using an offline multi-modal ML pipeline with three modules: branch diversity optimization, narration generation, and immersive audio navigation. It detects branching points, predicts preferred viewing directions, optimizes branch sets for spatial semantic and social diversity, and produces coherent audio descriptions with branch titles. Interaction uses subtle vibration and phone shaking to select options. Evaluation with 12 blind/low-vision participants shows significantly higher comprehension, agency, and usability versus pause-and-explore.
AI-written summary · about
chapters (8)
- 0:00Introduction to Branch Explorer
- 1:49Branching narratives concept
- 3:19User study insights
- 4:43System design and pipeline
- 6:50Evaluation and results
- 8:55Future directions
- 9:15Key takeaways and closing
- 9:38Q&A session
related talks
20mOutside-In: Visualizing Out-of-Sight Regions-of-Interest in a 360 Video Using Spatial ...
11mWatching 360° Videos Together
20mRich Representations of Visual Content for Screen Reader Users
21mImmersive Trip Reports
42mInteractive Data Analysis: Visualization and Beyond
17mAvoiding drill-down fallacies with VisPilot: assisted exploration of data subsets
20mVoyager 2: Augmenting Visual Analysis with Partial View Specifications
06mVirtual Tourism Through Mapping