GuideTWSI: A Diverse Tactile Walking Surface Indicator Dataset from Synthetic and Real-World Images for Blind and Low-Vision Navigation
Hochul Hwang, Soowan Yang, Nhat Hong Anh Nguyen, Parth Goel, Krisha Adhikari, Sunghoon Ivan Lee, Joydeep Biswas, Nicholas Giudice, Donghyun Kim
AI summary
Problem
Existing tactile paving datasets are geographically biased toward East Asian directional bars, lack robot-relevant viewpoints, and severely underrepresent truncated domes, limiting the reliability of autonomous mobility assistive robots.
Approach
The authors introduce GuideTWSI, a large-scale dataset combining a photorealistic Unreal Engine synthetic generation pipeline, curated open-source tactile data, and real-world quadruped robot-collected images for TWSI segmentation.
Key results
- GuideTWSI dataset comprising over 39K diverse synthetic, curated, and robot-collected samples
- Up to +29 mIoU improvement in truncated dome segmentation via synthetic augmentation
- 96.15% stop success rate with high repeatability in real-robot experiments
- Public release of dataset, code, and pretrained model weights
Why it matters
Enables safer, more reliable autonomous navigation assistance for blind and low-vision pedestrians by addressing critical geographic and viewpoint gaps in tactile indicator perception.
Abstract
Tactile Walking Surface Indicators (TWSIs) are safety-critical landmarks that blind and low-vision (BLV) pedes- trians use to locate crossings and hazard zones. From our observation sessions with BLV guide dog handlers, trainers, and an O&M specialist, we confirmed the critical importance of reliable and accurate TWSI segmentation for navigation assistance of BLV individuals. Achieving such reliability re- quires large-scale annotated data. However, TWSIs are severely underrepresented in existing urban perception datasets, and even existing dedicated paving datasets are limited: they lack robot-relevant viewpoints (e.g., egocentric or top-down) and are geographically biased toward East Asian directional bars— raised parallel strips used for continuous guidance along sidewalks. This narrow focus overlooks truncated domes— rows of round bumps used primarily in North America and Europe as detectable warnings at curbs, crossings, and platform edges. As a result, models trained only on bar-centric data struggle to generalize to dome-based warnings, leading to missed detections and false stops in safety-critical environments. We introduce GuideTWSI, the largest and most diverse TWSI dataset, which combines a photorealistic synthetic dataset, carefully curated open-source tactile data, and quadruped real-world data collected and annotated by the authors. No- tably, we developed an Unreal Engine–based synthetic data generation pipeline to obtain segmented, labeled data across diverse materials, lighting conditions, weather, and robot- relevant viewpoints. Extensive evaluations show that synthetic augmentation improves truncated dome segmentation across diverse state-of-the-art models, with gains of up to +29 mIoU points, and enhances cross-domain robustness. Moreover, real- robot experiments demonstrate accurate stoppings at truncated domes, with high repeatability and stop success rates (96.15%). The GuideTWSI dataset, model weights, and code are publicly released in https://guidedogrobot-tactile.github.io/.