Synthesizing Training Images for Boosting Human 3D Pose Estimation

doi:doi:10.57702/5p4893wl

Synthesizing Training Images for Boosting Human 3D Pose Estimation

Human 3D pose estimation from a single image is a challenging task with numerous applications. Convolutional Neural Networks (CNNs) have recently achieved superior performance on the task of 2D pose estimation from a single image, by training on images with 2D annotations collected by crowd sourcing. This suggests that similar success could be achieved for direct estimation of 3D poses.

BibTex:

@dataset{Wenzheng_Chen_and_Changhe_Tu_and_Huan_Wang_and_Dani_Lischinski_and_Yangyan_Li_and_Hao_Su_and_Zhenhua_Wang_and_Daniel_Cohen-Or_and_Baoquan_Chen_2024,
    abstract = {Human 3D pose estimation from a single image is a challenging task with numerous applications. Convolutional Neural Networks (CNNs) have recently achieved superior performance on the task of 2D pose estimation from a single image, by training on images with 2D annotations collected by crowd sourcing. This suggests that similar success could be achieved for direct estimation of 3D poses.},
    author = {Wenzheng Chen and Changhe Tu and Huan Wang and Dani Lischinski and Yangyan Li and Hao Su and Zhenhua Wang and Daniel Cohen-Or and Baoquan Chen},
    doi = {10.57702/5p4893wl},
    institution = {No Organization},
    keyword = {'3D Pose Estimation', 'Human Pose Estimation', 'Synthetic Data'},
    month = {dec},
    publisher = {TIB},
    title = {Synthesizing Training Images for Boosting Human 3D Pose Estimation},
    url = {https://service.tib.eu/ldmservice/dataset/synthesizing-training-images-for-boosting-human-3d-pose-estimation},
    year = {2024}
}