Info: E²DTC [5] is an end- to-end deep trajectory clustering framework that requires no manual feature extractions. It employs an encoder-decoder scheme to learn the representation of trajectories, and then uses a self-training method to jointly learn the cluster-oriented representation and perform the clustering analysis.
Info: E²DTC [5] is an end- to-end deep trajectory clustering framework that requires no manual feature extractions. It employs an encoder-decoder scheme to learn the representation of trajectories, and then uses a self-training method to jointly learn the cluster-oriented representation and perform the clustering analysis.