Advancements in Geometric Analysis and Machine Learning

The field of geometric analysis and machine learning is rapidly evolving, with a focus on developing innovative methods for predicting geometric deviations, mitigating biases in surgical operating rooms, and creating emergent morphogenesis via planar fabrication. Researchers are exploring new approaches to decouple geometry from optimization in 2D irregular cutting and packing problems, and investigating the use of breath as a biomarker for health monitoring. Additionally, there is a growing interest in developing robust methods for physical layer signal authentication, asymmetric stereo matching, and geometric shape assembly. Noteworthy papers include: Hybrid Machine Learning Framework for Predicting Geometric Deviations from 3D Surface Metrology, which achieved a prediction accuracy of 0.012 mm at a 95% confidence level. Mitigating Biases in Surgical Operating Rooms with Geometry, which demonstrated that geometric representations capture more meaningful biometric features than RGB models. Emergent morphogenesis via planar fabrication enabled by a reduced model of composites, which enabled efficient computational design and scalable manufacturing of 3D forms.

Sources

Hybrid Machine Learning Framework for Predicting Geometric Deviations from 3D Surface Metrology

Mitigating Biases in Surgical Operating Rooms with Geometry

Emergent morphogenesis via planar fabrication enabled by a reduced model of composites

Decoupling Geometry from Optimization in 2D Irregular Cutting and Packing Problems: an Open-Source Collision Detection Engine

AirSignatureDB: Exploring In-Air Signature Biometrics in the Wild and its Privacy Concerns

Breath as a biomarker: A survey of contact and contactless applications and approaches in respiratory monitoring

VeriPHY: Physical Layer Signal Authentication for Wireless Communication in 5G Environments

Iterative Volume Fusion for Asymmetric Stereo Matching

Combinative Matching for Geometric Shape Assembly

Enhanced Sparse Point Cloud Data Processing for Privacy-aware Human Action Recognition

Axis-level Symmetry Detection with Group-Equivariant Representation

Self-Supervised Stereo Matching with Multi-Baseline Contrastive Learning

Built with on top of