Markerless motion capture and sports biomechanics.
Same recipe — detect a pose, track it, classify it — but very different evidence. One 2026 paper grades a physical-therapy exercise against expert ground truth (RTMPose + GRU, ~96%); the other profiles a basketball move with no ground-truth labels, and honestly says so. What validates a pose pipeline isn’t the detector — it’s the ground truth.
You captured the motion — you’re not done. Two July 2026 papers trace the path from validated kinematics to a modeled injury number: an OpenCap markerless scoping review (51 studies), and a badminton-lunge ACL-load model (OpenSim + XGBoost, R² ≈ 0.91). Each inferential layer adds its own uncertainty.
Two 2026 papers take opposite bets: SoccerNet pools 427 teams and 1,129 submissions onto one shared benchmark, while a skateboarding study checks one commercial IMU against lab references — with mixed results.
Two reviews and one study show the distance between what AI-plus-wearables promises for injury prediction and what exists: a good classifier of expert judgment (0.86 AUC) — not a validated predictor of a future injury.
Cameras got cheap. Three 2026 studies — a 3D reconstruction pipeline, a systematic review, and one smartphone — all point the same way: global, gross measures are within reach; fine timing and joint angles are not, at any price point.
Wearable mocap used to mean a 17-sensor suit. Four 2026 studies ask the smallest set that still works: running needs 1–3 sensors, jump testing needs 1, base-running and hammer throw need more — and none was tested across all four.
CalTennis is the first large-scale sports benchmark for monocular-to-3D pose estimation: 40 players, 11M+ synchronized frames. The finding is a dissociation — relative joint angles pass, but depth and foot contact fail systematically, exactly where injury risk lives.
Four 2026 papers ask whether one phone camera can replace force plates, MoCap suits and EMG: OpenCap Monocular, BioHuman, From Pixels to Newtons, and Pose-to-Biomechanics — plus why upstream pose accuracy is the ceiling on all of them.
Three 2026 papers on getting biomechanics out of the lab: Ultra Diffusion Poser (full-body pose from IMUs + UWB, zero cameras), biomechanics-aware markerless hand capture, and a wearable study auditing the VR reality gap — plus the open question each one leaves behind.
Does single-leg jump strength predict soccer kick speed? In 6 amateur players, the dominant leg kicked harder (99 vs 85 km/h) — but the real surprise was that a stronger non-dominant support leg went with a faster dominant-leg kick.