Researchers have developed the In-Car Sign Language Corpus (ICSL), a new multi-modal dataset designed to advance sign language recognition (SLR) within confined spaces like vehicles. The dataset includes both high-precision motion capture data and real-world recordings from 2D cameras and 3D Time-of-Flight sensors, totaling over 1.5 million frames. This resource aims to improve public transport accessibility for the Deaf and Hard-of-Hearing community by enabling the training of robust SLR models for AI
IMPACT This dataset could enable more accessible public transport for the deaf and hard-of-hearing community by improving in-car communication.
RANK_REASON The cluster contains an academic paper describing a new dataset.
- Brazilian Sign Language
- City Law School
- deaf
- hard of hearing person
- In-Car Sign Language Corpus
- motion capture
- time of flight
AI-generated summary · Google Gemini · from 3 sources. How we write summaries →