resistance training

Abstract To compare the efficacy of lower-limb flywheel resistance training (FRT) versus conventional training (e.g., barbell and dumbbell exercises) on athletes’ sprinting and jumping abilities, and to identify optimal training dosages (frequency, duration, and volume). Four electronic databases were searched from inception to present. Following PICOS criteria, randomized and non-randomized controlled trials involving athletes and FRT interventions (≥ 3 weeks) were included. Data were synthesized using a random-effects model in R software, with Hedges’ g calculated as the effect size, and evidence certainty was assessed via the GRADE approach. Twelve studies (280 subjects) were included. Meta-analysis showed that FRT produced a small but significant improvement in jump height (g = 0.35, p < 0.05) and sprint time (g = − 0.32, p < 0.05) compared to conventional training. Subgroup analysis revealed a significant dose-response relationship: meaningful improvements required a duration of ≥ 8–9 weeks, ≥ 10 total sessions, specific volume threshold (i.e., sets × repetitions × number of sessions performed; ≥210 for jumping; ≥244 for sprinting). Furthermore, lower weekly frequencies (< 2 sessions/week) yielded larger effect sizes than higher frequencies. FRT, when implemented as a standalone modality, appears more effective than traditional training for enhancing sprinting and jumping abilities. To optimize adaptations, practitioners should implement longer interventions (≥ 8 weeks) with low weekly frequency (1–2 sessions) to balance high-intensity eccentric loading with neuromuscular recovery.

matched text: “resistance training”

Multi-view 3D human pose estimation drives automated sports and biomechanics evaluation but relies on fragile camera calibration.Conversely, recent parameter-free alternatives degrade severely under limited views.To overcome this, we propose the Global Attention Transformer (GAT).Our method emphasizes an extrinsic-free, singleframe architecture using joint-specific queries to model spatial and cross-view relationships without geometric triangulation.We employ a synthetic-to-real training strategy using extensive viewpoint randomization for robust generalization to uncalibrated setups.We also introduce an end-to-end biomechanical assessment pipeline extracting 2D keypoints via HRNet, lifting them to 3D via GAT, and evaluating weight training movements (correct or incorrect of squats and deadlifts) using a MiniROCKET classifier.GAT achieves a competitive Mean Per Joint Position Error (MPJPE) of 68.5 mm.Stress tests reveal strong generalization, maintaining 77.6 mm error under monocular views.Ultimately, the end-to-end pipeline achieves 95.31% classification accuracy on our collected dataset, indicating that GAT is a promising component for camera-agnostic weight training assessment under the evaluated conditions.

matched text: “weight training”