RadHARSimulator V2: Video to Doppler Generator

arXiv — cs.CV•Thursday, November 13, 2025 at 5:00:00 AM

RadHARSimulator V2, introduced on November 13, 2025, represents a breakthrough in radar-based human activity recognition (HAR) by allowing Doppler spectra generation from recorded video footage. This innovation addresses the limitations of existing software, which often relies on rigid models or motion-captured data, thus lacking flexibility. The simulator incorporates both computer vision and radar modules, employing methods like global nearest neighbor for object detection and Kalman filtering for smooth pose estimation. By utilizing advanced techniques such as the Savitzky-Golay method for pose interpolation and the short-time Fourier transform for Doppler time map generation, RadHARSimulator V2 enhances the accuracy and efficiency of HAR systems. This development is significant as it opens new avenues for research and application in fields requiring precise human activity monitoring.

— via World Pulse Now AI Editorial System

Read Original

Was this article worth reading? Share it

Recommended Readings

arXiv — cs.LGa day ago

A Survey of Cross-domain Graph Learning: Progress and Future Directions

NeutralArtificial Intelligence

Graph learning is essential for analyzing complex relationships in graph data, with applications in social, citation, and e-commerce networks. Despite the success of foundation models in computer vision (CV) and natural language processing (NLP), existing graph learning methods often lack generalization across domains. Cross-domain graph learning (CDGL) has emerged as a promising approach, aiming to create true graph foundation models. This survey reviews current CDGL research and proposes a taxonomy based on transferable knowledge types: structure-oriented, feature-oriented, and mixture-orien…

Read full article

via arXiv — cs.LG

arXiv — cs.LGa day ago

Optimizing Federated Learning by Entropy-Based Client Selection

PositiveArtificial Intelligence

The article discusses a novel approach to optimizing federated learning through a method called FedEntOpt. This technique addresses privacy concerns associated with centralized datasets by allowing multiple clients to collaboratively train a global deep learning model without exposing their data. FedEntOpt enhances model performance by selecting clients based on the entropy of the aggregated label distribution, effectively mitigating issues related to label skew. Experiments demonstrate that this method improves classification accuracy by up to 6% compared to existing algorithms.

Read full article

via arXiv — cs.LG

arXiv — cs.LG2 days ago

X-VMamba: Explainable Vision Mamba

PositiveArtificial Intelligence

The X-VMamba model introduces a controllability-based interpretability framework for State Space Models (SSMs), particularly the Mamba architecture. This framework aims to clarify how Vision SSMs process spatial information, which has been a challenge due to the absence of transparent mechanisms. The proposed methods include a Jacobian-based approach for any SSM architecture and a Gramian-based method for diagonal SSMs, both designed to enhance understanding of internal state dynamics while maintaining computational efficiency.

Read full article

via arXiv — cs.LG