Exploring Strategies for Modeling Sign Language Phonology
This work addresses sign language recognition for accessibility applications, but it is incremental as it builds on prior methods with specific improvements.
The paper tackled the problem of modeling sign language phonemes by exploring learning strategies like multi-task and curriculum learning, achieving an average accuracy of 87% across all phoneme types on the Sem-Lex Benchmark.
Like speech, signs are composed of discrete, recombinable features called phonemes. Prior work shows that models which can recognize phonemes are better at sign recognition, motivating deeper exploration into strategies for modeling sign language phonemes. In this work, we learn graph convolution networks to recognize the sixteen phoneme "types" found in ASL-LEX 2.0. Specifically, we explore how learning strategies like multi-task and curriculum learning can leverage mutually useful information between phoneme types to facilitate better modeling of sign language phonemes. Results on the Sem-Lex Benchmark show that curriculum learning yields an average accuracy of 87% across all phoneme types, outperforming fine-tuning and multi-task strategies for most phoneme types.