Zwei weitere Paper wurden auf dem 11th International Workshop on Computer Vision in Sports (CVsports) at CVPR 2025 akzeptiert: "Leveraging Anthropometric Measurements to Improve Human Mesh Estimation and Ensure Consistent Body Shapes" von Katja Ludwig, Julian Lorenz, Daniel Kienzle, Tuan Bui & Rainer Lienhart sowie "Efficient 2D to Full 3D Human Pose Uplifting including Joint Rotations" von Katja Ludwig, Yuliia Oksymets, Robin Sch?n, Daniel Kienzle & Rainer Lienhart.

Im ersten Paper beschreiben die Autoren Unstimmigkeiten in bisherigen Methoden zur Sch?tzung von 3D Meshes von Personen - wird so eine Methode auf ein Video einer Person angewendet, ?ndert sich die K?rperform der Person von Frame zu Frame. Die Autoren stellen deshalb die Methode A2B vor, bei der man anthropometrische Ma?e verwendet, um die K?rperform eines Menschen codiert im K?rpermodell SMPL-X zu erhalten. Diese feste K?rperform kann nun für alle Videos dieser Person verwendet werden. Au?erdem wird in diesem Paper beschrieben, wie ein 3D Posensch?tzer kombiniert mit inverser Kinematik und A2B für Sportdatens?tze deutlich bessere Ergebnisse liefert als aktuelle Mesh-Sch?tzer.

Im zweiten Paper wird diese Idee aufgegriffen und verbessert. Die inverse Kinematik ist ein iterativer Algorithmus, der pro Frame extra ausgeführt werden muss und damit sehr viel Zeit ben?tigt. Die Autoren analysieren deshalb eine Erweiterung des 3D Posensch?tzers "Uplift and Upsample" (Einfalt et al.) um die Sch?tzung von 3D Gelenkrotationen, die in der Basisversion nicht enthalten sind. Auf diese Weise kann das System 150 mal schneller gemacht werden.

?

Leveraging Anthropometric Measurements to Improve Human Mesh Estimation and Ensure Consistent Body Shapes:

The basic body shape (i.e., the body shape in T-pose) of a person does not change within a single video. However, most SOTA human mesh estimation (HME) models output a slightly different, thus inconsistent basic body shape for each video frame. Furthermore, we find that SOTA 3D human pose estimation (HPE) models outperform HME models regarding the precision of the estimated 3D keypoint positions. We solve the problem of inconsistent body shapes by leveraging anthropometric measurements like taken by tailors from humans. We create a model called A2B that converts given anthropometric measurements to basic body shape parameters of human mesh models. We obtain superior and consistent human meshes by combining the A2B model results with the keypoints of 3D HPE models using inverse kinematics. We evaluate our approach on challenging datasets like ASPset or fit3D, where we can lower the MPJPE by over 30 mm compared to SOTA HME models. Further, replacing estimates of the body shape parameters from existing HME models with A2B results not only increases the performance of these HME models, but also guarantees consistent body shapes.

?

Efficient 2D to Full 3D Human Pose Uplifting including Joint Rotations:

In sports analytics, accurately capturing both the 3D locations and rotations of body joints is essential for understanding an athlete's biomechanics. While Human Mesh Recovery (HMR) models can estimate joint rotations, they often exhibit lower accuracy in joint localization compared to 3D Human Pose Estimation (HPE) models. Recent work addressed this limitation by combining a 3D HPE model with inverse kinematics (IK) to estimate both joint locations and rotations. However, IK is computationally expensive. To overcome this, we propose a novel 2D-to-3D uplifting model that directly estimates 3D human poses, including joint rotations, in a single forward pass. We investigate multiple rotation representations, loss functions, and training strategies — both with and without access to ground truth rotations. Our models achieve state-of-the-art accuracy in rotation estimation, are 150 times faster than the IK-based approach, and surpass HMR models in joint localization precision.