← Back to all news
arXiv Computer Vision

Training-Free Hold-Usage Detection in Sport Climbing with Foundation Pose Models

The paper presents a training‑free method for detecting which holds a climber uses in sport climbing videos by leveraging a frozen foundation pose model (Sapiens) that provides fingertip and toe keypoints. Using a simple proximity test, mutual exclusion, and a temporal‑persistence rule, the approach achieves high F_1 scores (up to 90.2%) on the Way Up dataset without any climbing‑specific training, outperforming repurposed pose pipelines. The resulting automatic predictions enable accurate coaching statistics, such as climb time and pace, with Pearson correlations of 1.00 and 0.94 respectively.

Related stories

Hugging Face Trending Papers
3d ago

One Sensor, Whole Body - 3D Body Pose from a Single Consumer Earbud IMU

The study investigates how well a single consumer earbud IMU can estimate 3D body pose and whether adding foot IMUs improves accuracy. Using a multimodal capture pipeline with RGB‑D video, an AirPods head IMU, and Striv insole IMUs, the authors benchmark pose estimation across various motions and train recurrent models (IMUPoser and MobilePoser). Results show that a head IMU alone achieves 79.0 mm rigid‑MPJPE and 0.809 macro‑F1 for foot contact, while adding foot IMUs does not significantly improve pose and can even degrade performance due to insole orientation errors.

fine-tuningmultimodalbenchmarkssafety
More like this →
arXiv Computer Vision
Sep 3

MuyBridge: Mobile Human Center-of-Mass Estimation from Monocular Video via Sparse Fusion

MuyBridge is an on-device system that estimates an athlete’s segmental center of mass (CoM) trajectory from a single phone camera video stream. It combines a compact 2D pose network with a distilled monocular depth network, fusing their outputs through anatomical and physical priors to produce metric CoM estimates without requiring 3D or task‑specific supervision. On the AthletePose3D dataset, MuyBridge achieves 33–41 mm vertical CoM error and 2.3–6.6 % absolute‑relative range error, delivering CoM estimates at 63 FPS on an iPhone 15 with asynchronous depth updates.

By Aidan Bradshaw, Marco Giordano, David Rode, Andreas Habersack, Elif Basokur, Annika Kruse, Markus Tilp, Michele Magno, Peter Wolf, Luca Benini, Christoph Leitner
More like this →
arXiv Machine Learning
Sep 24

When Adaptation Hurts: Split Sensitivity and Person-Level Negative Transfer in Federated Wearable Onboarding

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.

By Rahil Aftab, Vineet Kumar Rakesh, Soumya Mazumdar, Tapas Samanta
benchmarkssafety
More like this →
arXiv Computer Vision
Sep 3

Cross-Model Distillation of a Human-Pose Foundation Model from Unannotated Infant Video for Markerless 3D Pose Estimation

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.

By R. James Cotton, Divya Joshi, Colleen Peyton
computer-visionfine-tuningefficiency
More like this →
arXiv Computer Vision
Sep 3

Video-Based Palm-Vein Authentication under Challenging Conditions

The paper introduces the Columbia University Palm‑vein (CUP) dataset, the first public video‑based palm‑vein dataset that captures palms under four surface conditions—clean, warm, wet, and dirty—along with physiological and demographic metadata. Twenty‑one recognizers are benchmarked on CUP, revealing that models performing well on clean palms lose most accuracy on dirty palms, with mean EER roughly quadrupling. The authors propose a lightweight design that fuses global cosine similarity with a saliency‑steered region‑level optimal transport, achieving state‑of‑the‑art performance across all surfaces while reducing parameters and computational cost, and they identify demographic gaps in warm‑condition performance.

By Xiaofeng Yan, Kechen Liu, Abhilash Venkatesh, Cathy Zhang, Xia Zhou, Salvatore Stolfo
benchmarks
More like this →