XmoPipe: A Pipeline for Large-Scale In-the-Wild Human Motion Dataset Construction
Provides a scalable, low-cost alternative to marker-based motion capture for building diverse human motion datasets, benefiting motion analysis and generation research.
XmoPipe constructs large-scale in-the-wild human motion datasets from online videos, enabling training of motion models that perform comparably to those trained on traditional motion capture datasets with strong cross-dataset generalization.
Large-scale human motion datasets are essential for training robust motion models for analysis, synthesis, and understanding. While marker-based motion capture provides precise data, it is costly and limited in scale and diversity. Recent advances in monocular motion capture and video-language understanding open the way to extract plausible motion from unconstrained online videos. We present a scalable pipeline for constructing in-the-wild human motion datasets. From a few keywords, the system retrieves videos, extracts 3D body and facial motion, and generates high-level textual descriptions. The pipeline is flexible, enabling targeted collection of various motions, multi-person interactions, or expressive behaviors. We demonstrate its quality by training motion reconstruction and motion generation models, showing performance comparable to models trained on traditional motion capture datasets and strong cross-dataset generalization.