Physical Geometry
3D/4D reconstruction, articulated and deformable scene understanding, geometry-aware world models, and physically grounded view synthesis.
NeurIPS 2026 workshop proposal
Physical World AI: Geometry, Characteristics, and Multimodal Sensing
Final proposal details reflected from the PDF
Acronym: PhysWorldAI
Name: Physical World AI: Geometry, Characteristics, and Multimodal Sensing
Modern computer vision and embodied AI systems must act in the physical world, not merely describe it. The workshop frames physical world understanding around geometry, physical characteristics, and multimodal sensing, connecting vision, robotics, graphics, haptics, audio, and physics-based simulation.
Submission, review, and publication policy
From structure to material properties and non-RGB sensing
3D/4D reconstruction, articulated and deformable scene understanding, geometry-aware world models, and physically grounded view synthesis.
Material and physical property estimation, mass, friction, stiffness, elasticity, deformability, affordances, contact-rich interaction, differentiable simulation, and generative models of physical dynamics.
Multimodal sensing and fusion with tactile, force/torque, proprioceptive, RF, audio, depth, IMU, and event-based signals.
Embodied world models, robot manipulation, sim-to-real transfer, multimodal simulators, benchmarks, datasets, evaluation protocols, and responsible deployment.
Sponsor commitments listed in the proposal
CyberBrain and 2077AI are listed as confirmed sponsors. The proposal names an Overall Best Paper Award and a Best Student Paper Award, with support also planned for student travel grants for underrepresented attendees and on-site logistics.