The novel subject triplet loss provides the best performance overall, and all personalized deep embeddings out-perform our baseline personalized engineered feature embedding and an impersonal fully convolutional neural network classifier.
Human Action Recognition is an important task of Human Robot Interaction as cooperation between robots and humans requires that artificial agents recognise complex cues from the environment.
In addition we visualise pressure mat data to explore the potential of the sensor to capture exercise performance quality.
We conclude that Human Activity Recognition systems should be evaluated by Subject Cross Validation, and that overlapping windows are not worth their extra computational cost.
Time Series Classification (TSC) problems are encountered in many real life data mining tasks ranging from medicine and security to human activity recognition and food safety.
Human activity recognition is typically addressed by detecting key concepts like global and local motion, features related to object classes present in the scene, as well as features related to the global context.
Inspired by this, we conduct an extensive set of experiments that analyze different sample generation processes and validation protocols to indicate the vulnerable points in human activity recognition based on wearable sensor data.
We propose to use the various slices (such as $x-y$, $x-t$, and $y-t$) of the DVS video as a feature map for HAR and denote them as Motion Maps.
No spatial coherence is forced on the glimpse locations, which gives the module liberty to explore different points at each frame and better optimize the process of scrutinizing visual information.
#5 best model for Skeleton Based Action Recognition on N-UCLA