Fingerspelling PoseNet: Enhancing Fingerspelling Translation with Pose-Based Transformer Models

We address the task of American Sign Language finger-spelling translation using videos in the wild. We exploit advances in more accurate hand pose estimation and pro-pose a novel architecture that leverages the transformer based encoder-decoder model enabling seamless contextual word translation. The translation model is augmented by a novel loss term that accurately predicts the length of the finger-spelled word, benefiting both training and inference. We also propose a novel two-stage inference approach that re-ranks the hypotheses using the language model ca-pabilities of the decoder. Through extensive experiments, we demonstrate that our proposed method outperforms the state-of-the-art models on ChicagoFSWild and ChicagoF-SWild+ achieving more than 10% relative improvement in performance. Our findings highlight the effectiveness of our approach and its potential to advance fingerspelling recognition in sign language translation. Code is also available at https://github.com/pooya£ayyaz/Fingerspe11ing-PoseNet.