Context

EgoMotion is the 2nd edition of EgoMotion Workshop for in-the-wild human motion modeling using egocentric multi-modal data from wearable devices. We focus on topics and sub-fields in motion tracking, synthesis, and understanding. These research topics have garnered increasing interest and momentum thanks to recent advances in wearable egocentric sensors. Specific topics include:

• Body tracking, synthesis, and activity understanding from egocentric/exocentric cameras
• Body tracking, synthesis, and activity understanding from non-visual wearable sensors, e.g., inertial measurement units (IMUs), electromagnetic (EM) sensors, ultra-wideband (UWB) sensors, radar, and audio.
• Applications of egocentric motion for character animation, simulation, robotic learning, etc.

In addition to algorithms, the workshop aims to promote recent open-source projects to encourage and accelerate research in this field. To this end, we invite researchers to present recent large-scale mo- tion datasets and associated challenges, such as GORP, Nymeria, EgoLife, HP-EPIC, EgoBody, etc. The workshop will also discuss state-of-the-art human motion modeling libraries such as Meta Momentum, present research hardware platforms such as Project Aria, and organize live demos.