Collecting complex activity datasets in highly rich networked sensor environments

Abstract

Abstract-We deployed 72 sensors of 10 modalities in 15 wireless and wired networked sensor systems in the environment, in objects, and on the body to create a sensor-rich environment for the machine recognition of human activities. We acquired data from 12 subjects performing morning activities, yielding over 25 hours of sensor data. We report the number of activity occurrences observed during post-processing, and estimate that over 13000 and 14000 object and environment interactions occurred. We describe the networked sensor setup and the methodology for data acquisition, synchronization and curation. We report on the challenges and outline lessons learned and best practice for similar large scale deployments of heterogeneous networked sensor systems. We evaluate data acquisition quality for on-body and object integrated wireless sensors; there is less than 2.5% packet loss after tuning. We outline our use of the dataset to develop new sensor network self-organization principles and machine learning techniques for activity recognition in opportunistic sensor configurations. Eventually this dataset will be made public.

Detecting and Rectifying Anomalies in Body Sensor Networks

“…The second dataset, termed the Opportunity dataset, contains data for daily home activities in a breakfast scenario [13]. Five IMU sensor nodes are located on back, lower Figure 2.…”

Unsupervised adaptation for acceleration-based activity recognition: robustness to sensor displacement and rotation

“…The data is a subset of a larger recording performed in a rich-sensor environment [16,20]. It corresponds to a daily living scenario where the subject performs a morning activities.…”

Identifying Important Action Primitives for High Level Activity Recognition

“…We used the data sets that have been collected in the EU project OPPORTUNITY [11]. The data set about the naturalistic human activities was collected in a sensor rich environment: a room simulating a studio flat with kitchen, deckchair, and outdoor access where subjects performed daily morning activities.…”

“…For example, "hand cutting the bread movement" is an important action primitive as it is giving a clear indication that subject is preparing a sandwich. We used the annotations of the EU project OPPORTU-NITY data set [11] that are based on the recordings of the proceedings of data collection activity as action primitives. These annotations include body movement primitives, like walk, sit, and stand, arm movement primitives, like reach, release, and cut, and object or environment usage primitives, like use glass, move chair, and open fridge door.…”