{"type":"ScholarlyArticle","title":"Human Motion Capture from Loose and Sparse Inertial Sensors with Garment-aware Diffusion Models","authors":[{"name":"Andela Ilic","slug":"andela-ilic","given":"Andela","family":"Ilic","affiliation":""},{"name":"Jiaxi Jiang","slug":"jiaxi-jiang","given":"Jiaxi","family":"Jiang","affiliation":""},{"name":"Paul Streli","slug":"paul-streli","given":"Paul","family":"Streli","affiliation":""},{"name":"Xintong Liu","slug":"xintong-liu","given":"Xintong","family":"Liu","affiliation":""},{"name":"Christian Holz","slug":"christian-holz","given":"Christian","family":"Holz","affiliation":""}],"abstract":"Motion capture using sparse inertial sensors has shown great promise due to its portability and lack of occlusion issues compared to camera-based tracking. \r\nExisting approaches typically assume that IMU sensors are tightly attached to the human body. \r\nHowever, this assumption often does not hold in real-world scenarios. \r\nIn this paper, we present a new task of full-body human pose estimation using sparse, loosely attached IMU sensors.\r\nTo solve this task, we simulate IMU recordings from an existing garment-aware human motion dataset.\r\nWe developed transformer-based diffusion models to synthesize loose IMU data and estimate human poses based on this challenging loose IMU data.\r\nIn addition, we show that incorporating garment-related parameters while training the model on simulated loose data effectively maintains expressiveness and enhances the ability to capture variations introduced by looser or tighter garments. \r\nExperiments show that our proposed diffusion methods trained on simulated and synthetic data outperformed the state-of-the-art methods quantitatively and qualitatively, opening up a promising direction for future research.","doi":"10.24963/ijcai.2025/135","url":"https://ijcai.org/proceedings/2025/135","pdf":"https://ijcai.org/proceedings/2025/0135.pdf","markdown":"https://ijcai.org/proceedings/2025/135.md","bibtex":"https://ijcai.org/proceedings/2025/bibtex/135","volume":{"slug":"2025","title":"Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2025","url":"https://ijcai.org/proceedings/2025/","year":2025,"location":"Montreal, Canada","dates":"16-22 August 2025","editors":"James Kwok","isbn":"978-1-956792-06-5","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2025"},"section":"Computer Vision","track":"Main Track","number":135,"pages":{"first":1206,"last":1214},"keywords":["Computer Vision: CV: Biometrics, face, gesture and pose recognition","Computer Vision: CV: Motion and tracking"],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}