Human Motion Challenges in Real-World and Clinical Settings

MoCha @ ECCV 2026

Malmö, Sweden

Room: TBD

Date: TBDD

+ Benchmark and Challenge on Parkinsonian Gait

A focused forum on human motion understanding under realistic constraints, bridging computer vision, biomechanics, digital health, and clinically grounded evaluation.

View Challenge

About

Human motion understanding is central to computer vision, yet many models still degrade when motion patterns diverge from the curated distributions seen in standard benchmarks. These gaps become even more important in clinical settings, where subtle movement deviations can carry meaningful diagnostic value.

This workshop creates a meeting point for researchers working on human-centric vision, long-term motion understanding, robust evaluation, and clinically grounded modeling. The goal is to push the field beyond controlled benchmarks toward methods that remain reliable under real-world constraints.

Topics include:

3D pose and mesh reconstruction in unconstrained environments
Long-term and longitudinal human motion understanding
Efficient parametric human representations
Motion representation learning and generative modeling
Privacy-preserving human motion modeling in real-world contexts
Learning under domain shift and cross-site generalization
Fairness, reproducibility, generalization, and bias in human motion models
Benchmark design for real-world human motion understanding
Gait analysis and kinematic modeling
Musculoskeletal and muscle-actuated humanoid modeling

Benchmark & Challenge

The workshop is paired with a benchmark and challenge on Parkinsonian gait built on CARE-PD, a large harmonized dataset of anonymized SMPL mesh gait sequences aggregated across nine cohorts and clinical centers.

The proposed challenge uses CARE-PD as training data and evaluates performance on harmonized test sequences from three additional unseen sites. This setup is designed to reward methods that remain robust to real-world distribution shift and clinically realistic variability.

Challenge important dates:

Start of the public benchmark: TBD
Close of the public benchmark: TBD

(The benchmark will continue to be available for evaluation after this date, but the results will not be considered for the workshop.)

Participate

Invited Speakers

Gerard Pons-Moll

University of Tübingen

Siyu Tang

ETH Zürich

Fabien Baradel

NAVER LABS Europe

Javier Romero

Reality Labs Research, Meta

R. James Cotton

Northwestern University & Shirley Ryan AbilityLab

Ehsan Adeli

Stanford University

Call for Papers

We invite submissions on human motion understanding under realistic and clinically grounded conditions, spanning computer vision, biomechanics, digital health, and robust evaluation. The workshop welcomes work on both methodological advances and deployment-aware analysis.

Submissions aligned with the workshop theme will be peer reviewed by a program committee composed of the organizers and researchers with expertise in human-centric vision. A best paper will be selected and highlighted during the workshop.

Relevant topics include 3D pose and mesh reconstruction in the wild, motion representation learning, domain shift and cross-site generalization, gait analysis, motion biomarkers, fairness and bias in motion models, benchmark design, and privacy-preserving human representations for healthcare settings.