Understandings of human activity through wearable sensors will enable the next-generation human-oriented computing. However, most of researches on the activity recognition so far are based on small number of test subjects, and not well adapted for real world applications. To overcome the situation, we have started a project named "HASC Challenge" to collect a large scale human activity corpus. By the end of 2010, by the collaboration of 20 teams, more than 6700 accelerometer data with 540 subjects have been collected through our project. We also developed a tool named "HASC Tool" for management, evaluation and collection of the large number of activity sensor data.

Human activity recognition through the wearable sensor will enable a next-generation human-oriented biquitous computing. However, most of research on human activity recognition so far is based on small number of subjects, and non-public data. To overcome the situation, we have gathered 4897 accelerometer data with 116 subjects and compose them as HASC2011corpus. In the field of pattern recognition, it is very important to evaluate and to improve the recognition methods by using the same dataset as a common ground. We make the HASC2011corpus into public for the research community to use it as a common ground of the Human Activity Recognition. We also show several facts and results of obtained from the corpus.UbiComp \u2711 Proceedings of the 13th international conference on Ubiquitous computing, September 17-21, 2011, Beijing, Chin

Pedestrian ow analysis using Wi-Fi packet sensors can easily collect the movement trajectories of individuals, has a high-resolution sensing ability, and is well suited to indoor operations. Thus it is regarded as one of the promising sensing technologies for Smart Cities targeting a variety of applications such as disaster prevention, the planning of commercial and barrier-free facilities, and monitoring people with disabilities. However, because Wi-Fi packet sensors cannot detect devices that have disabled their Wi-Fi functionality or people with no Wi-Fi devices, they can only probabilistically estimate the number of people in a target area. We conducted a two-month field experiment in a shopping mall, and derived a coefficient for estimating the actual number of people in the area by comparing the number of recognized devices in the mall, as counted by Wi-Fi packet sensors, with data from motion detectors placed at all of the mall's entrances and exits. In the calculation process, we separated the collected data into weekdays and holidays by considering the visiting patterns, and calculated an estimation coefficient α for the actual number of people. This gives the least absolute error between the observed values given by the Wi-Fi packet sensors and hourly data of motion detectors over the course of a week. The accuracy of α is evaluated by comparing known data from the motion detectors with the estimated people numbers calculated from the Wi-Fi packet sensor data. As a result, we demonstrate that the maximum error in the number of people in the mall for each hour, accumulated over a day, is less than 7.9%.