Presentation

<
Neural Particle Automata: Learning Self-Organizing Particle Dynamics
SessionGenerative 3D (2)
DescriptionNeural Particle Automata extend Neural Cellular Automata from fixed grids to dynamic particles. Each particle carries a continuous position and internal state, updated by a shared neural rule with SPH-based local perception. Backed by optimized CUDA kernels, NPA learn scalable self-organizing dynamics for morphogenesis, particle texture synthesis, and point-cloud classification.
Authors
Ehsan Pajouheshgar
EPFL
Hyunsoo Kim
Korea Advanced Institute of Science and Technology (KAIST)
Sabine Süsstrunk
EPFL
Wenzel Jakob
EPFL
Jinah Park
Korea Advanced Institute of Science and Technology (KAIST)
Event Type
Technical Paper
TimeTuesday, 21 July 20269:30am - 9:40am PDT
LocationRoom 403 B
Digital Library PDF
Session Time & Location
Sunday, 19 July 20266:00pm - 8:45pm PDTHall K
Tuesday, 21 July 20269:00am - 10:30am PDTRoom 403 B
Keywords
Geometry
Modeling
Registration Categories
Full Conference Supporter
Full Conference
Next PresentationNext PresentationPixTex: Consistent 3D Texturing via Pixel-Space Multi-View Diffusion
Similar Presentations
Neural Cellular Automata: From Cells to Pixels
Generative Modeling with Orbit-Space Particle Flow Matching
Learning View-Dependent Splatting Kernels