Search papers, labs, and topics across Lattice.
This paper introduces semantic motion anchors, a novel approach that bridges the gap between spoken text and gesture by capturing both the physical form and communicative intent of gestures. By discretizing 3D gestures into motion primitives and grounding them in transcripts, the method enhances retrieval performance, achieving an 8.2% improvement in text-to-gesture retrieval over traditional direct text-motion baselines. The results indicate that this approach not only improves retrieval metrics but also yields gestures that are more semantically meaningful and better suited for conveying intent in downstream applications.
Semantic motion anchors enable a dramatic 8.2% boost in text-to-gesture retrieval accuracy by grounding gesture motion in communicative intent, not just kinematics.
Learning a shared representation between spoken text and gesture is central to co-speech gesture retrieval, synthesis, and understanding, but remains challenging for semantically meaningful gestures whose communicative intent is not captured by motion alone. Direct contrastive alignment between transcripts and continuous motion embeddings often overemphasizes low-level kinematics and misses the symbolic content of semantic gestures. We propose semantic motion anchors, natural-language abstractions of gesture motion capturing physical form and communicative intent. Our method discretizes 3D gestures into body-hand motion primitives, verbalizes them into structured descriptions, and grounds them in the transcript to provide auxiliary contrastive supervision. On BEAT2, our method improves text-to-gesture R@1 by 8.2% over a direct text-motion baseline and outperforms prior retrieval approaches on text to gesture and gesture to text retrieval directions. Beyond aggregate retrieval metrics, semantic motion anchor supervision helps retrieve gestures that are semantically meaningful for the spoken query, rather than defaulting to generic motion patterns. A downstream retrieval-augmented gesture generation study showed that users significantly preferred gestures retrieved by our approach over a retrieval-augmented generation baseline, demonstrating that semantically grounded retrieval translates to gestures that better convey communicative intent in downstream generation.
