The Robot Report

Humanoid says KinetIQ Ascend reinforcement learning approaches human-level dexterity

Humanoid says its KinetIQ Ascend approach can reach 99.9% manipulation reliability at human speed and beyond for industrial ...
IEEE

Privacy-Preserving Facial and Action Recognition Using Federated Deep Learning on Edge Devices

Abstract: With the growing demand for real-time visual recognition on edge devices, preserving user privacy has become a critical challenge. Facial and action recognition systems must ensure data ...
GitHub

VLAC: A Vision-Language-Action-Critic Model for Robotic Real-World Reinforcement Learning

VLAC is a general-purpose pair-wise critic and manipulation model which designed for real world robot reinforcement learning and data refinement. It provides robust evaluation capabilities for task ...
Frontiers

Editorial: Advancing human wellbeing: environment-focused AI technologies

Artificial intelligence is increasingly influencing modern research and is becoming ever more positioned at the intersection of scientific and technological innovation, environmental sustainability, ...
IEEE

Lightweight Multimodal Feature Fusion and Spatiotemporal Learning for Human Action Recognition on Edge Devices

Abstract: Human action recognition (HAR) remains a challenging topic in computer vision, attracting extensive research for applications in surveillance, sports, and human-computer interaction.