The workshop is motivated by the observations that people spend a significant part of their daily lives performing a variety of activities in the physical world—travelling to places (including commuting to/from work using public or private transport), dwelling and engaging in various activities at various locations (e.g., exercising in the gym, eating at restaurants and food courts) , interacting with various physical objects and artefacts (e.g., touching or picking up products at a retail store, or browsing through books and magazines at a library), being subject to various audiovisual stimuli (e.g., listening to announcements at transit hubs, watching advertisements on public displays or movies on TV) and interacting with other people (in groups, as part of crowds or one-on-one). These activities and interactions contain a wealth of information about user behavior, preferences, attitudes and interests, that, if harnessed, can benefit both users and consumer-facing businesses.
While research has been underway in utilizing various sensing and analytics tools to capture and annotate such behavior (e.g., profile smoking episodes using wearable devices or monitor consumer reactions to advertising content via video analysis), the vast majority of such research focuses on exploring individual sensing techniques targeted at specific activities, and is scattered across various academic forums. The goal of this workshop is to offer a unified forum to explore both (a) the technologies (current and emerging) that can enable unobtrusive capture of such individual and collective physical world behavior, and (b) the real-world commercial applications and services that leverage upon such understanding of physical world behavior.
In particular, the workshop solicits submissions that relate to three distinct questions:
- What types of technologies (e.g., use of existing or new mobile/wearable sensors, use of passive RF monitoring) can capture different facets of physical world behavior?
- How actionable are the insights on physical behavior context that are generated via analytics over such captured data streams—in other words, what types of accuracy, false positives/negatives result from the use of various technologies, and what analytics tools can enhance these performance metrics?
- What early examples of commercial usage of physical analytics are emerging across different vertical domains, and what bottlenecks remain to be overcome for successful utilization of such analytics-generated insights?
Topics of interest (NOT an exhaustive list):
Please submit you papers through EasyChair: https://easychair.org/conferences/?conf=wpa16 .
The deadline for submissions is
|08:45 - 09:00||Workshop Opening and Welcome|
|09:00 - 10:00||Keynote Talk by Prof. Kyle Jamieson|
|10:00 - 10:30||Tea/Coffee Break|
|10:30 - 12:00||Session 1: People Analytics|
|12:00 - 13:00||Lunch break|
|13:00 - 15:00||Session 2 : Space Analytics|
|15:00 - 15:30||Tea/Coffee Break|
|15:30 - 17:00||Session 3: Social Analytics|
|17:00 - 17:45||Panel and Closing|
Keynote Talk: "Tracking Mobiles, Objects, and People in the Wireless Internet of Things" by Prof. Kyle Jamieson, Princeton University.
Abstract: Phased array signal processing has long been employed outdoors in radar, underwater in sonar, and underground in seismic monitoring. Today, it has the potential to revolutionize the Internet of Things (IoT) by giving us the ability to track every one of the billions of IoT devices indoors, and meet their exploding bandwidth requirements. But to make the shift to indoor wireless networks, it must cope with strong multipath radio reflections, packetized data transmissions, and commodity hardware platforms. In this talk I will describe two relevant systems through the lens of system-building and experimentation. First, I will describe an indoor location system that uses solely the existing Wi-Fi infrastructure to achieve a median location accuracy of 23 centimeters, and sub-second response time, allowing Wi-Fi-enabled devices to be tracked in real-time. Next, I will present a system that can order RFID-tagged books placed on a library shelf 3-5 cm from each other, detecting any misshelved books. Finally, I will conclude with ongoing work in fine-grained motion tracking.
Speaker Bio: Kyle Jamieson is an Assistant Professor in the Department of Computer Science at Princeton University and Honorary Reader at University College London. His research interests are in building wirelessly networked systems for the real world that cut across the boundaries of digital communications and networking. He received the B.S., M.Eng., and Ph.D. (2008) degrees in Computer Science from the Massachusetts Institute of Technology. He then received a Starting Investigator fellowship from the European Research Council in 2011, Best Paper awards at USENIX 2013 and CoNEXT 2014, and a Google Faculty Research Award in 2015. He regularly serves on the program committees of the ACM MobiCom, USENIX NSDI, and ACM SIGCOMM conferences.
Paper Session 1: People Analytics
- MyDrive: Drive Behavior Analytics Method and Platform. Tanushree Banerjee, Arijit Chowdhury, Tapas Chakravarty (Tata Consultancy Services)
- AnnoTainted: Automating Physical Activity Ground Truth Collection Using Smartphones. Rahul Majethia (Shiv Nadar University), Akshit Singhal (University of Texas), Lakshmi Manasa K, Kunchay Sahiti, Shubhangi Kishore, Vijay Nandwani (Shiv Nadar University)
- DXTK: Exploring Deep Learning on Mobile and Embedded Platforms with the DeepX Toolkit. Nicholas D. Lane, Sourav Bhattacharya Claudio Forlivesi, Akhil Mathur, Fahim Kawsar (Bell Labs, Nokia)
Paper Session 2: Space Analytics
- MobiCamp: A Campus-wide Testbed for Studying Mobile Physical Activities. Mengyu Zhou, Kaixin Sui, Minghua Ma, Youjian Zhao, Dan Pei (Tsinghua University), Thomas Moscibroda (Microsoft Research)
- Next Generation Physical Analytics for Digital Signage. Mateusz Mikusz, Anastasios Noulas, Nigel Davies, Sarah Clinch, Adrian Friday (Lancaster University)
- Small Scale Deployment of Seat Occupancy Detectors. Nguyen Huy Hoang Huy (Singapore Management University), Gihan Hettiarachchi (University of Moratuwa), Youngki Lee, Rajesh Krishna Balan (Singapore Management University)
- Margdarshak: A Mobile Data Analytics based Commute Time Estimator cum Route Recommender. Rohit Verma, Aviral Shrivastava, Sandip Chakraborty, Bivas Mitra (Indian Institute of Technology)
Paper Session 3: Social Analytics
- A Trained-once Crowd Counting Method Using Differential WiFi Channel State Information. Simone Di Domenico, Mauro De Sanctis, Ernestina Cianca, Giuseppe Bianchi (University of Roma Tor Vergata)
- Capturing Personal and Crowd Behavior with Wi-Fi Analytics. Utku Gunay Acer, Geert Vanderhulst, Afra Masshadi, Aidan Boran, Claudio Forlivesi (Bell Labs, Nokia), Philipp M. Scholl (Albert- Ludwigs-Universität Freidburg), Fahim Kawsar (Bell Labs, Nokia)
- Fusing WiFi and Video Sensing for Accurate Group Detection in Indoor Spaces. Kasthuri Jayarajah (Singapore Management University), Zaman Lantra (University of Moratuwa), Archan Misra (Singapore Management University)