NCSTR: Node-Centric Decoupled Spatio-Temporal Reasoning for Video-based Human Pose Estimation

Video-based human pose estimation remains challenged by motion blur, occlusion, and complex spatiotemporal dynamics. Existing methods often rely on heatmaps or implicit spatio-temporal feature aggregation, which limits joint topology expressiveness and weakens cross-frame consistency. To address these problems, we propose a novel node-centric framework that explicitly integrates visual, temporal, and structural reasoning for accurate pose estimation. First, we design a visuo-temporal velocity-based joint embedding that fuses sub-pixel joint cues and inter-frame motion to build appearance- and motion-aware representations. Then, we introduce an attention-driven pose-query encoder, which applies attention over joint-wise heatmaps and frame-wise features to map the joint representations into a pose-aware node space, generating image-conditioned joint-aware node embeddings. Building upon these node embeddings, we propose a dual-branch decoupled spatio-temporal attention graph that models temporal propagation and spatial constraint reasoning in specialized local and global branches. Finally, a node-space expert fusion module is proposed to adaptively fuse the complementary outputs from both branches, integrating local and global cues for final joint predictions. Extensive experiments on three widely used video pose benchmarks demonstrate that our method outperforms state-of-the-art methods. The results highlight the value of explicit node-centric reasoning, offering a new perspective for advancing video-based human pose estimation.

Wearable slot antenna at 2.45 GHz for off-body radiation: analysis of efficiency, frequency shift and body absorption

The interaction of body worn antennas with the human body causes a significant decrease in the antenna efficiency and a shift in the resonant frequency. A resonant slot in a small conductive box placed on the body has been shown to reduce these effects. The specific absorption rate (SAR) is less than international health standards for most wearable antennas due to the small transmitter power. This paper reports the linear relationship between the power absorbed by biological tissues at different locations on the body, and the radiation efficiency based on numerical modeling (r = 0.99). While the -10 dB bandwidth of the antenna remains constant and equal to 12.5%, the maximum frequency shift occurs when the antenna is close to the elbow (6.61%) and on the thigh (5.86%). The smallest change was found on the torso (4.21%). Participants with body-mass index (BMI) between 17 and 29 kg/m2 took part in experimental measurements, where the maximum frequency shift was 2.51%. Measurements show better agreement with simulations on the upper arm. These experimental results demonstrate that the BMI for each individual has little effect on the performance of the antenna.

RF Energy Absorption in Human Bodies Due to Wearable Antennas in the 2.4 GHz Frequency Band

Human exposure to electromagnetic fields produced by two wearable antennas operating in the 2.4 GHz frequency band was assessed by computational tools. Both antennas were designed to be attached to the skin, but they were intended for different applications. The first antenna was designed for off-body applications, i.e. to communicate with a device placed outside the body, while the second antenna model was optimized to communicate with a device located inside the body. The power absorption in human tissues was determined at several locations of adult male and female body models. The maximum specific absorption rate (SAR) value obtained with the off-body antenna was found on the torso of the woman model and was equal to 0.037 W/kg at 2.45 GHz. SAR levels increased significantly for the antenna transmitting inside the body. In this case, SAR values ranged between 0.23 and 0.45 W/kg at the same body location. The power absorbed in different body tissues and total power absorbed in the body were also calculated; the maximum total power absorbed was equal to 5.2 mW for an antenna input power equal to 10 mW.