TY - JOUR
T1 - Learning to recognize human activities using soft labels
AU - Hu, Ninghang
AU - Englebienne, Gwenn
AU - Lou, Zhongyu
AU - Kröse, Ben
PY - 2016/10/16
Y1 - 2016/10/16
N2 - Human activity recognition system is of great importance in robot-care scenarios. Typically, training such a system requires activity labels to be both completely and accurately annotated. In this paper, we go beyond such restriction and propose a learning method that allow labels to be incomplete and uncertain. We introduce the idea of soft labels which allows annotators to assign multiple, and weighted labels to data segments. This is very useful in many situations, e.g., when the labels are uncertain, when part of the labels are missing, or when multiple annotators assign inconsistent labels. We formulate the activity recognition task as a sequential labeling problem. Latent variables are embedded in the model in order to exploit sub-level semantics for better estimation. We propose a max-margin framework which incorporate soft labels for learning the model parameters. The model is evaluated on two challenging datasets. To simulate the uncertainty in data annotation, we randomly change the labels for transition segments. The results show significant improvement over the state-of-the-art approach.
AB - Human activity recognition system is of great importance in robot-care scenarios. Typically, training such a system requires activity labels to be both completely and accurately annotated. In this paper, we go beyond such restriction and propose a learning method that allow labels to be incomplete and uncertain. We introduce the idea of soft labels which allows annotators to assign multiple, and weighted labels to data segments. This is very useful in many situations, e.g., when the labels are uncertain, when part of the labels are missing, or when multiple annotators assign inconsistent labels. We formulate the activity recognition task as a sequential labeling problem. Latent variables are embedded in the model in order to exploit sub-level semantics for better estimation. We propose a max-margin framework which incorporate soft labels for learning the model parameters. The model is evaluated on two challenging datasets. To simulate the uncertainty in data annotation, we randomly change the labels for transition segments. The results show significant improvement over the state-of-the-art approach.
U2 - 10.1109/TPAMI.2016.2621761
DO - 10.1109/TPAMI.2016.2621761
M3 - Article
VL - 39
SP - 1973
EP - 1984
JO - IEEE Transactions on Pattern Analysis and Machine Intelligence
JF - IEEE Transactions on Pattern Analysis and Machine Intelligence
IS - 10
ER -