Fuzzy Logic based CNN Ensembling for 3D Action Recognition

Implementation of the paper - Fuzzy Integral based CNN Classifier Fusion for 3D Skeleton Action Recognition - IEEE Transactions on Circuits and Systems for Video Technology in TensorFlow

Uses CNNs trained on complementary features from human skeleton sequence, ensembled at inference time for classification. Predicts actions from 3D video feed.

Ensembling strategy utilizes the Choquet Integral, inspired by the principles of Fuzzy Logic.

Citation

If this paper and/or this code is useful for your research, please consider citing us -

@ARTICLE{9177170,
  author={A. {Banerjee} and P. K. {Singh} and R. {Sarkar}},
  journal={IEEE Transactions on Circuits and Systems for Video Technology}, 
  title={Fuzzy Integral based CNN Classifier Fusion for 3D Skeleton Action Recognition}, 
  year={2020},
  volume={},
  number={},
  pages={1-1},}