These tools and metrics are designed to help AI actors develop and use trustworthy AI systems and applications that respect human rights and are fair, transparent, explainable, robust, secure and safe.
Mean Per Joint Position Error (MPJPE)
Mean Per Joint Position Error (MPJPE) is a common metric used to evaluate the performance of human pose estimation algorithms. It measures the average distance between the predicted joints of a human skeleton and the ground truth joints in a given dataset. The lower the MPJPE, the better the performance of the algorithm.
In the context of 3D human pose estimation, the MPJPE is calculated as the mean Euclidean distance between the predicted 3D joint locations and the corresponding ground truth joint locations. This metric is used to evaluate how accurately the algorithm is able to predict the 3D pose of a person in an image or video.
Trustworthy AI Relevance
This metric addresses Robustness and Transparency by quantifying relevant system properties. Robustness: MPJPE measures the Euclidean distance between predicted and ground-truth joint positions, producing a direct quantitative measure of prediction accuracy. By computing MPJPE across different subsets (noise levels, occlusion conditions, different domains/OOD inputs, or across time), practitioners can evaluate resilience to adverse conditions and distribution shifts, detect performance degradation, and compare model variants for reliability.
Related use cases :
3D Hand Reconstruction via Aggregating Intra and Inter Graphs Guided by Prior Knowledge for Hand-Object Interaction Scenario
Uploaded on Mar 15, 2024HOISDF: Constraining 3D Hand-Object Pose Estimation with Global Signed Distance Fields
Uploaded on Mar 15, 2024About the metric
You can click on the links to see the associated metrics
Objective(s):
Purpose(s):
Target sector(s):
Lifecycle stage(s):
Target users:
