Reproduce “Proximity3D: Shape from Capacitive Proximity on Sensing Manifold” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Hao Chen, Chenming Wu, Chun Ping Lam, Xiangjia Chen, Guoxin Fang, Charlie C. L. Wang, Yeung Yam, Juncong Lin and 1 more
Proximity3D enables the recovery of complete 3D object geometry from sparse, multi-view capacitive proximity fields detected on a curved textile sensing manifold. The framework, incorporating Manifold Sensing Attention, achieves a 90% grasp planning success rate by providing precise pre-contact shape information, supported by a surrogate model that generates realistic training data.
Key insights
- Explicitly incorporating local electrode geometry and tangent-frame information via Manifold Sensing Attention (MSA) is crucial for accurate interpretation of sparse capacitive proximity fields on curved surfaces.
- A hybrid data generation strategy, combining large-scale FEM simulations with fine-tuning on limited real sensor data, effectively overcomes physical data collection limitations.
- Reconstructing detailed 3D object shapes from multi-view capacitive proximity fields provides a precise understanding of the near-field, greatly enhancing robotic manipulation capabilities.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.