EgoForce: Robust Online Egocentric Motion Reconstruction via Diffusion Forcing
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911679434981376 |
|---|---|
| author | Hwang, Inwoo Lim, Donggeun Jang, Hojun Kim, Young Min |
| author_facet | Hwang, Inwoo Lim, Donggeun Jang, Hojun Kim, Young Min |
| contents | With recent advances in embodied agents and AR devices, egocentric observations are readily available as input for real-world interactive online applications. However, egocentric viewpoints can only sporadically observe hands, in addition to the estimated head trajectory. We propose EgoForce, an online framework for reconstructing long-term full-body motion from noisy egocentric input. While existing generative frameworks can robustly handle noisy and sparse measurements, they assume a fixed-length observation window is available and are thus not suitable for real-time applications. Faster inference often relies on autoregressive prediction, sacrificing robustness. In contrast, we adopt a diffusion-based method with a temporally asymmetric noise schedule inspired by Diffusion Forcing. Specifically, our approach models temporally evolving uncertainty and incrementally denoises states as new streaming observations arrive. Combined with a noise-robust imputation strategy, EgoForce progressively generates stable and coherent full-body motion under strict causal constraints. Experiments demonstrate that our online framework outperforms existing online and offline methods, enabling long-horizon, full-body motion reconstruction in challenging egocentric scenarios. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_13041 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | EgoForce: Robust Online Egocentric Motion Reconstruction via Diffusion Forcing Hwang, Inwoo Lim, Donggeun Jang, Hojun Kim, Young Min Computer Vision and Pattern Recognition With recent advances in embodied agents and AR devices, egocentric observations are readily available as input for real-world interactive online applications. However, egocentric viewpoints can only sporadically observe hands, in addition to the estimated head trajectory. We propose EgoForce, an online framework for reconstructing long-term full-body motion from noisy egocentric input. While existing generative frameworks can robustly handle noisy and sparse measurements, they assume a fixed-length observation window is available and are thus not suitable for real-time applications. Faster inference often relies on autoregressive prediction, sacrificing robustness. In contrast, we adopt a diffusion-based method with a temporally asymmetric noise schedule inspired by Diffusion Forcing. Specifically, our approach models temporally evolving uncertainty and incrementally denoises states as new streaming observations arrive. Combined with a noise-robust imputation strategy, EgoForce progressively generates stable and coherent full-body motion under strict causal constraints. Experiments demonstrate that our online framework outperforms existing online and offline methods, enabling long-horizon, full-body motion reconstruction in challenging egocentric scenarios. |
| title | EgoForce: Robust Online Egocentric Motion Reconstruction via Diffusion Forcing |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2605.13041 |