aiqbench analysis
FALLNEST addresses the critical challenge of training deep models for fall detection using scarce, imbalanced IMU data by introducing a novel dataset condensation technique. This method distills large corpora into tiny, class-balanced synthetic sets without relying on re-sampling, enabling high-accuracy TinyCNNs to train on minimal data while preserving subject-wise split integrity. The approach represents a significant advance in data-centric AI for safety-critical applications where labeled event data is inherently rare.
FALLNEST: Dual-Domain Dataset Condensation of Imbalanced IMU Data for Wearable Fall Detection
arXiv:2610.18260v1 Announce Type: cross Abstract: Wearable fall detection is hampered by rare fall events, scarce labeled IMU data, and the cost of training deep models on large, imbalanced corpora. Existing methods either rely on hand-crafted features with shallow classifiers, which limits learning from raw signals, or train deep networks on full datasets that require re-sampling to counter class skew and demand substantial data and compute. To address this, we propose a novel Fall detection framework via Normalized-gradient and Embedding matched Synthetic Time-frequency condensation, dubbed FALLNEST, a data-centric approach that distills large, imbalanced IMU datasets into tiny, class-balanced synthetic training sets. Building on dual-domain time-series condensation (CondTSC), FALLNEST introduces a fall-specific 7-channel input with accelerometer vector magnitude, class-balanced K-Means initialization, and normalized gradient matching with gradient clipping, combined with MMD-based embedding alignment and multi-view Fourier-domain augmentation, so that class balance arises from the condensed set itself rather than from re-sampling. We evaluate on SisFall, KFall, and MobiFall using strict subject-wise splits and naturally imbalanced test sets, reporting accuracy, Macro-F1, and PR-AUC over five seeds. A 102K-parameter TinyCNN trained on only 32 to 146 synthetic windows (about only 1% of the training windows) reaches 93.5-95.8% mean accuracy and 0.89-0.98 PR-AUC, with standard deviations of 1.0-1.7% in accuracy, demonstrating the efficacy of our approach.
Abstract:Wearable fall detection is hampered by rare fall events, scarce labeled IMU data, and the cost of training deep models on large, imbalanced corpora. Existing methods either rely on hand-crafted features with shallow classifiers, which limits learning from raw signals, or train deep networks on full datasets that require re-sampling to counter class skew and demand substantial data and compute. To address this, we propose a novel Fall detection framework via Normalized-gradient and Embedding matched Synthetic Time-frequency condensation, dubbed FALLNEST, a data-centric approach that distills large, imbalanced IMU datasets into tiny, class-balanced synthetic training sets. Building on dual-domain time-series condensation (CondTSC), FALLNEST introduces a fall-specific 7-channel input with accelerometer vector magnitude, class-balanced K-Means initialization, and normalized gradient matching with gradient clipping, combined with MMD-based embedding alignment and multi-view Fourier-domain augmentation, so that class balance arises from the condensed set itself rather than from re-sampling. We evaluate on SisFall, KFall, and MobiFall using strict subject-wise splits and naturally imbalanced test sets, reporting accuracy, Macro-F1, and PR-AUC over five seeds. A 102K-parameter TinyCNN trained on only 32 to 146 synthetic windows (about only 1% of the training windows) reaches 93.5-95.8% mean accuracy and 0.89-0.98 PR-AUC, with standard deviations of 1.0-1.7% in accuracy, demonstrating the efficacy of our approach.
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Signal Processing (eess.SP)
arXiv:2610.18260 [cs.CV]
(or arXiv:2610.18260v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2610.18260 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 309 words
Related coverage
VSMP-IMU: Video-Grounded Semantic Motion Programs for Sensor-Aware Synthetic IMU Generation
arXiv · Aug 7, 2026
Active Learning with Foundation Model Priors: Efficient Learning under Class Imbalance
arXiv · Jun 9, 2026
Where Detectors Fail: Closing the Tail-Domain Gap with Expert-Guided Mutual Distillation
arXiv · Jul 30, 2026
Nemotron 3.5 Content Safety: Customizable Multimodal Safety for Global Enterprise AI
Hugging Face · Jun 4, 2026