MySign: A High-Fidelity Motion-Capture Dataset for 3D Sign Generation in Bahasa Isyarat Malaysia

Progress in fine-grained 3D sign generation is constrained by the scarcity of high-fidelity 3D sign language datasets, a limitation that is particularly severe for under-resourced languages. We introduce MySign, the first motion-capture dataset for isolated signs in Bahasa Isyarat Malaysia (BIM), comprising 5,000 3D samples covering 1,000 lexical items (glosses), produced by five Deaf native signers. Unlike prior datasets that rely on SMPL-X parameters regressed from video (pseudo-labels)--often affected by depth ambiguity, inaccurate finger articulation, and temporal jitter--MySign provides ground-truth SMPL-X sequences captured with a motion-capture setup tailored for sign language. This yields higher-fidelity motion, particularly in fine-grained hand and finger articulation. We establish MySign as a challenging benchmark for gloss-conditioned 3D sign generation: state-of-the-art models produce visually plausible outputs but fall short of ground-truth fidelity, particularly in fine-grained articulation, revealing persistent gaps in modeling articulated sign motion. In contrast, recognition baselines consistently achieve high lexical accuracy, indicating that the dataset reliably supports supervised learning. We anticipate that MySign will enable more faithful 3D sign generation and support finer-grained analysis of sign articulation. The dataset is available at https://huggingface.co/datasets/mysigner/MySign.