Sep 2026· Italian National Conference on Sensors· Vol 26· 0 citations· 44 references
Medicine
TL;DR
A three-stage Adaptive CNN-Enhanced Scale-fusion Network (ACESNet) for six-channel accelerometer–gyroscope activity recognition supports a favorable performance–efficiency trade-off relative to the selected baselines under the stated protocol.
Abstract
Deep learning has achieved notable success in sensor-based human activity recognition (SHAR), yet wearable inertial signals contain activity patterns at different temporal scales while practical deployment imposes strict computational constraints. This paper proposes a three-stage Adaptive CNN-Enhanced Scale-fusion Network (ACESNet) for six-channel accelerometer–gyroscope activity recognition. The Adaptive Kernel Encoder dynamically combines temporal kernels of different sizes, while the CNN-Enhanced Scale-fusion design combines multi-scale temporal mixing, explicit channel mixing and a parallel local CNN path without materializing a global attention matrix. The main benchmark uses a fixed stratified 70/15/15 window-level partition (split seed 42), and all methods in the main comparison are repeated over five training seeds. On REALWORLD, MotionSense, UCI-HAR and SHL, ACESNet achieves accuracies of 95.37 ± 0.09%, 99.33 ± 0.12%, 98.01 ± 0.23% and 93.49 ± 0.10%, with Macro-F1 scores of 95.59 ± 0.09%, 99.12 ± 0.15%, 98.16 ± 0.21% and 94.03 ± 0.08%, respectively. A separate official subject-independent UCI-HAR evaluation yields 94.01 ± 0.84% accuracy and 94.11 ± 0.88% Macro-F1. The adaptive-kernel selector has a mean normalized entropy of 0.9114, providing evidence against severe single-branch collapse rather than strong kernel selectivity. On a Raspberry Pi 5 (Raspberry Pi Ltd., Cambridge, UK), ACESNet requires 8.731 MFLOPs and 1.606 ± 0.014 ms per window in FP32 ONNX Runtime 1.18.1. These results support a favorable performance–efficiency trade-off relative to the selected baselines under the stated protocol.
A resource-constrained CNN-GRU hybrid model for HAR on the WISDM dataset that uses convolutional layers for spatial learning and gated recurrent units (GRU) for sequence learning, enabling real-time HAR on edge devices.
H. S. Ganesha, K. G. Harsha, Roshini P et al.· International journal of com...· 0 citations
A distributed HAR framework in which five wearable sensors are associated with local embedded nodes that perform acquisition, windowing, preprocessing, and convolutional neural network–long short-term memory (CNN–LSTM) inference.
Jhonathan L. Rivas-Caicedo, Laura Saldaña-Aristizábal, Kevin Niño-Tejada et al.· Electronics· 0 citations
These results demonstrate the models’ respective strengths in capturing long-range temporal dependencies, bidirectional temporal patterns, and short-term temporal dynamics with cross-channel interactions in complex motion patterns.
Wearable inertial measurement units (IMUs) provide a practical and privacy-preserving sensing modality for lower-limb activity recognition in assistive robotics, rehabilitation monitoring, and movement analysis. However, many IMUbased human activity recognition methods model multichannel sensor streams mainly as flat t...
Alvarado Morales Lisbeth Katherine, Qi-Fei Wu, Guoyu Zuo et al.· IEEE/ASME International Conf...· 0 citations
Findings provide a leakage-resistant but selection-sensitive benchmark for subject-independent DeepConvLSTM evaluation on WISDM and showed an observed class-level trade-off relative to last-timestep pooling in the evaluated comparison rather than a general architectural advantage.
Fakhrul Zidan Nurrohman, C. Dewa· bit-Tech· 0 citations
Smartphone-based human activity recognition (HAR) requires discriminative temporal modeling and careful assessment of computational cost. This study evaluates a Physics-Informed ConvFormer with State-Space Gating for six-class HAR on the WISDM v1.1 accelerometer dataset. The model combines raw acceleration with determi...
Yan-Cheng Pan, Zi-Xin Zhao, Ying-Wei Xu· Italian National Conference...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.