Spatio-temporal evaluation of white cane maneuvering based on a single video
Abstract
This study presents a method for evaluating the three-dimensional posture of white cane walking from a single camera observation and proposes an approach to evaluate the quality of white cane maneuvering through spatio-temporal analysis of the estimated results. The camera is set at a certain place to capture all the walking path. To achieve 3D posture estimation, we introduce an extended skeletal model that incorporates the white cane as part of the human skeleton. White cane maneuvering is a critical skill directly linked to the safety of visually impaired pedestrians, and certain postures are regarded as appropriate for effective and safe use. In this work, we compare walking with and without conscious adoption of the appropriate posture by applying the extended skeletal model and conducting spatio-temporal analysis of the resulting motion. For validation, we implemented a system based on the proposed method and analyzed factors such as cane swing width, the state of the stepping foot, and the position of the cane relative to the forward walking direction, thereby demonstrating the effectiveness of the approach.
Fields of Study & Subjects
Human Pose and Action Recognition
Assistive Technology in Communication and Mobility
Hand Gesture Recognition Systems
Publication details
- Journal / source
- International Workshop on Advanced Imaging Technology (IWAIT)
- Pages
- 113–113
- Publication date
- 27 February 2026
- Language
- EN