Tutorials
Learning to Understand Body Language from Flight through Robust 3D Avatar Placing
arXiv:2607.27865v1 Announce Type: new Abstract: Perceiving human motion and intent at long range is a prerequisite for socially intelligent aerial robots, yet the data to learn it barely exists. We in
arXiv:2607.27865v1 Announce Type: new Abstract: Perceiving human motion and intent at long range is a prerequisite for socially intelligent aerial robots, yet the data to learn it barely exists. We introduce Drones2BodyLanguage, a dataset grounding human motion in real UAV footage: avatars manifesting ten communicative intents are placed into unmodified 4K drone scenes with metrically correct position, scale and orientation, maintained over hundreds of frames of camera motion. Enabling it is a lightweight geometric world model of the local scene - semantically selected anchors lifted to 3D through streaming monocular depth - in which a placement point is predicted as an affine anchor combination with provably rigid-invariant weights, and re-rendered under an SVD-fitted ground rotation. Across twelve architectures on scene- and motion-disjoint splits, training on placed data lifts mean intent accuracy by a wide margin for real, retargeted and generated motion alike, with gains confirmed on two in-the-wild scenes.
Related
- Counterfactual Motion Reliability Learning for Robust UAV Tracking
- X-Diffusion: Training Diffusion Policies on Cross-Embodiment Human Demonstrations
- Learning Long-term Motion Embeddings for Efficient Kinematics Generation
- Cross-Domain Human Action Recognition from Multiview Motion and Textual Descriptions
Source: arXiv cs.CV | 2026-07-31