The study evaluates how well a surface‑EMG gesture recogniser trained on intact‑limb data transfers to transradial amputees. Zero‑shot transfer fails; the model needs a few labelled repetitions from the new user to outperform a per‑user classifier, achieving a macro‑F1 of 0.779 versus 0.589. Training on a larger pool of intact subjects, or combining intact and amputee data, yields the best cross‑population performance.
arXiv:2607. 27565v1 Announce Type: new Abstract: Pattern-recognition control promises a myoelectric prosthesis that responds to many intended gestures rather than one or two, but the promise has stayed in the laboratory.
Pattern-recognition control promises a myoelectric prosthesis that responds to many intended gestures rather than one or two, but the promise has stayed in the laboratory. A recogniser trained on one person rarely transfers to the next, and useful performance usually demands a fresh round of labelled calibration from the end user.
arXiv:2607. 27568v1 Announce Type: new Abstract: Recognition accuracy obtained during a recording session does not persist when a user puts on the electrodes again after the electrodes had previously been removed.
arXiv:2609.23352v1 Announce Type: new Abstract: Bimanual interaction produces complementary tactile views of the same physical process, yet existing tactile representation learning largely models the...
The paper evaluates six onboarding strategies for federated wearable models on five datasets using a leakage‑controlled protocol that fixes source checkpoints and separates calibration from evaluation. Results show that while average accuracy is high, person‑level performance can drop significantly, with some methods causing negative transfer for certain users. The study highlights that mean accuracy alone is insufficient and provides an auditable benchmark and failure map for future development.
arXiv:2607. 04820v1 Announce Type: new Abstract: Decoding hand kinematics from surface electromyography (EMG) is a core challenge in wearable biosignal processing with clinical relevance for prosthetic control and motor rehabilitation.
The study audits the impact of exercise‑specific joint selection on skeleton‑based correctness classification using 1,057 repetitions from ten REHAB24‑6 subjects. It finds that the manual‑subset kNN gain varies from 0.055 for pooled out‑of‑fold AUROC to 0.020 for equal‑weight within‑person AUROC, with both intervals including zero. Across 1,000 dimension‑matched random maps, 14 match or exceed the manual pooled result, while 145 do so when bilateral structure and trunk inclusion are also matched; RBF‑SVM shows a positive within‑person gain, whereas logistic regression and a random‑convolution comparator show negative gains under that estimand.
The paper investigates how to personalize a cross-user surface electromyography (sEMG) encoder when only a few calibration repetitions are available. Four methods—prototypical adaptation, linear probes, scaled fine‑tuning, and full fine‑tuning—were evaluated across 77 subjects on two databases. Full fine‑tuning consistently achieved the highest accuracy, but a gradient‑free prototypical rule captured 52–78 % of the benefit without per‑user weight copies, enabling quick donning‑time personalization.
arXiv:2607. 24519v2 Announce Type: replace Abstract: Pretrained EEG foundation models are proposed for clinical decoding, but whether reported gains transfer across populations or survive negative controls is unclear.
The study investigates whether Low‑Rank Adaptation (LoRA) can adapt three pretrained EEG foundation models—LaBraM‑base, REVE‑base, and REVE‑large—for binary left‑ vs. right‑hand motor imagery decoding in stroke patients. Using subject‑wise five‑fold cross‑validation on the PhysioNet EEG Motor Movement/Imagery Dataset and a binary subset of the UET175 stroke dataset, LoRA significantly improved accuracy for LaBraM‑base (0.822) and REVE‑base (0.957) on the healthy cohort, but only REVE‑base LoRA achieved high performance (0.847±0.194) on stroke data, with a best mean accuracy of 0.952 in leave‑one‑subject‑out evaluation. The results demonstrate that healthy‑benchmark performance does not guarantee transfer to stroke EEG, highlighting the need for target‑domain adaptation and subject‑level assessment in rehabilitation BCIs.
arXiv:2609.37297v1 Announce Type: new Abstract: Cross-skeleton motion generation trains generative models to carry action structure and motion intention from one body to another. Yet a target motion...
The paper presents a method for improving markerless 3D pose estimation in infants by cross‑model distillation. Using unannotated infant video, a frozen Sapiens 2 pose model provides dense pseudo‑labels that guide fine‑tuning of the SAM 3D Body model. On a held‑out dataset of eleven infants, the fine‑tuned model shows significant gains in 2D keypoint accuracy and 3D joint error compared to the original SAM 3D Body model.