AIGaitor is the first claimed end-to-end on-device monocular motion-capture and deep-learning gait analysis pipeline demonstrated on consumer smartphones.
Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
citation-role summary
background 1
baseline 1
method 1
citation-polarity summary
years
2026 8representative citing papers
HARMES is the first large-scale dataset to combine wrist IMU, environmental, and audio sensors for recognizing 15 household activities across over 80 hours of data from 20 participants.
Gated Multi-modal Fusion reaches 0.82 macro F1 on HARMES, beating the concatenation baseline of 0.76 by 6 points under leave-one-participant-out evaluation.
A large-scale benchmark of 17 WHAR models across 30 datasets finds predictive performance has plateaued while efficiency favors compact neural models and random forests on the Pareto frontier.
AnyMo pre-trains a graph encoder on physics-simulated multi-placement IMU data and aligns full-body motion tokens with LLMs to enable zero-shot activity recognition, retrieval, and captioning across unseen datasets and setups.
Hierarchical CNN-LSTM plus vision transformer detects tremor from raw time-domain kinematic data across nine body parts with average F1 of 0.765 and attention-based explanations.
Markerless pipeline estimates Rodda-Graham knee (R²=0.80) and ankle (R²=0.57) z-scores from single-view videos in 152 children with 60 diagnoses, achieving AUROC=0.88 for excess knee flexion screening.
TRACE improves activity recognition accuracy and temporal coherence in smart homes by integrating multi-source sensor evidence with contextual priors.
citing papers explorer
-
AIGaitor: Privacy-preserving and cloud-free motion analysis for everyone, using edge computing
AIGaitor is the first claimed end-to-end on-device monocular motion-capture and deep-learning gait analysis pipeline demonstrated on consumer smartphones.
-
HARMES: A Multi-Modal Dataset for Wearable Human Activity Recognition with Motion, Environmental Sensing and Sound
HARMES is the first large-scale dataset to combine wrist IMU, environmental, and audio sensors for recognizing 15 household activities across over 80 hours of data from 20 participants.
-
A Comparison of Fusion Techniques for Multi-Modal Human Activity Recognition on the HARMES Dataset
Gated Multi-modal Fusion reaches 0.82 macro F1 on HARMES, beating the concatenation baseline of 0.76 by 6 points under leave-one-participant-out evaluation.
-
WHAR Arena: Benchmarking the State of the Art in Efficient Wearable Human Activity Recognition
A large-scale benchmark of 17 WHAR models across 30 datasets finds predictive performance has plateaued while efficiency favors compact neural models and random forests on the Pareto frontier.
-
AnyMo: Geometry-Aware Setup-Agnostic Modeling of Human Motion in the Wild
AnyMo pre-trains a graph encoder on physics-simulated multi-placement IMU data and aligns full-body motion tokens with LLMs to enable zero-shot activity recognition, retrieval, and captioning across unseen datasets and setups.
-
An explainable hierarchical self attention-based approach for tremor detection in the time domain
Hierarchical CNN-LSTM plus vision transformer detects tremor from raw time-domain kinematic data across nine body parts with average F1 of 0.765 and attention-based explanations.
-
Quantifying Rodda and Graham Gait Classification from 3D Markerless Kinematics derived from a Single-view Video in a Heterogeneous Pediatric Clinical Cohort
Markerless pipeline estimates Rodda-Graham knee (R²=0.80) and ankle (R²=0.57) z-scores from single-view videos in 152 children with 60 diagnoses, achieving AUROC=0.88 for excess knee flexion screening.
-
TRACE: Temporal Reasoning over Context and Evidence for Activity Recognition in Smart Homes
TRACE improves activity recognition accuracy and temporal coherence in smart homes by integrating multi-source sensor evidence with contextual priors.