Deep Imputation for Skeleton data (DISK) for behavioral science
Open Access
- 4 December 2025
- journal article
- Published by Springer Nature in Nature Methods
- Vol. 23 (1) , 236-247
Abstract
Pose estimation methods and motion capture systems have opened doors to quantitative measurements of animal kinematics. While animal behavior experiments are expensive and complex, tracking errors sometimes make large portions of the experimental data unusable. Here our deep learning method, Deep Imputation for Skeleton data (DISK), uncovers dependencies between keypoints and their dynamics to impute missing tracking data without the help of any manual annotations. We demonstrate the utility and performance of DISK on seven animal skeletons including multi-animal setups. The imputed recordings allow us to detect more episodes of motion, such as steps, and obtain more statistically robust results when comparing these episodes between experimental conditions. In addition, by learning to impute the missing content, DISK learns meaningful representations of the data capturing, for example, underlying actions. This stand-alone imputation package, available at https://github.com/bozeklab/DISK.git/ , is applicable to outputs of tracking methods (marker-based or markerless) and allows for varied types of downstream analysis.Keywords
All Related Versions
This publication has 42 references indexed in Scilit:
- Differences between lower extremity joint running kinetics captured by marker-based and markerless systems were speed dependentJournal of Sport and Health Science, 2024
- SAITS: Self-attention-based imputation for time seriesExpert Systems with Applications, 2023
- Animal Pose Tracking: 3D Multimodal Dataset and Token-based Pose OptimizationInternational Journal of Computer Vision, 2022
- Self-supervised 3D human pose estimation from videoNeurocomputing, 2022
- Graph transformer network with temporal kernel attention for skeleton-based action recognitionKnowledge-Based Systems, 2022
- Distinct movement patterns generate stages of spider web buildingCurrent Biology, 2021
- Anipose: A toolkit for robust markerless 3D pose estimationCell Reports, 2021
- Decoupled Spatial-Temporal Attention Network for Skeleton-Based Action-Gesture Recognition2021
- DeepAR: Probabilistic forecasting with autoregressive recurrent networksInternational Journal of Forecasting, 2020
- Forecasting stock markets using wavelet transforms and recurrent neural networks: An integrated system based on artificial bee colony algorithmApplied Soft Computing, 2011