2nd International Workshop on 
Behavior analysis and Recognition for knowledge Discovery (BiRD 2023)

In conjunction with the IEEE International Conference on Pervasive Computing and Communications (PerCom 2023), March 13-17, 2023 in Atlanta, USA.

About
BiRD workshop aims to provide a forum for presenting and discussing research on the recognition and understanding of behavior data collected from various sources, such as humans, animals, and automobiles, towards scientific discovery (e.g., ecology and medical science) and/or actual business applications (e.g., industry, sports, healthcare, and stockbreeding). In addition, we plan to invite biologists and neuroscientists who have collected and analyzed behavior data of various animals. While the biologists and neuroscientists have massive amounts of behavior data, they have limited knowledge on state-of-the-art behavior analysis techniques. Therefore, this workshop provides a good opportunity for the PerCom researchers to apply the developed state-of-the-art methods to actual problems.

Important dates

Workshop schedule 

TBD

Call for papers (PDF version)

International Workshop on Behavior analysis and Recognition for knowledge Discovery (BiRD 2023)
Due to the recent advances in sensing technologies, massive amounts of activity/behavior data are being collected from humans, animals, and automobiles. For example, activity and trajectory data of humans and automobiles can be easily collected by smart devices such as smartphones. In addition, small GPS and acceleration loggers enable us to collect behavior data from animals such as birds, bears, cattle, and turtles to better understand the ecology of the animals. Furthermore, activity recognition and indoor positioning techniques using the sensor data of humans have been actively studied in the PerCom community. However, computational methods for knowledge discovery in the sensory data and/or recognized behavior data have not yet been fully explored. Moreover, many pervasive computing researchers are eager to apply their state-of-the-art techniques to actual problems and contribute to scientific discovery. However, there are few academic conferences that mainly focus on knowledge discovery techniques for behavior/activity data, making it difficult for the researchers to develop methods that are actually used/needed.    

This workshop solicits papers on the recognition and understanding of behavior data collected from various sources, such as humans, animals, and automobiles, towards scientific discovery (e.g., ecology and medical science) and/or actual business applications (e.g., industry, sports, healthcare, and stockbreeding). Because the organizers of this workshop are core members of an interdisciplinary project of engineer, computer science, biology, and neuroscience on behavior understanding (https://bio-navigation.jp/en/) and one of the organizers is a biologist specialized in seabirds, we plan to invite biologists and neuroscientists who have collected and analyzed behavior data of various animals. While the biologists and neuroscientists have massive amounts of behavior data, they have limited knowledge on state-of-the-art behavior analysis techniques. Therefore, this workshop provides a good opportunity for the PerCom researchers to apply the developed state-of-the-art methods to actual problems. The topics of interest include, but are not limited to:

Submission

Authors are invited to submit full papers that are unpublished and not under review elsewhere. In addition, we solicit poster papers that mainly focus on real data and discuss experience/limitations of conventional analysis methods applied to the data. The papers for the workshops should be at most 6 pages in the IEEE template. Poster papers (2 pages) must be stated as such.

Papers can be submitted via the following site.
Submission site
Each accepted workshop paper requires a full PerCom registration (no registration is available for workshops only). For preparation of the camera ready paper, please refer to:
TBD
Workshop papers will be included and indexed in the IEEE digital libraries (Xplore).

OpenPack Challenge 2022

OpenPack Challenge 2022 hosted by BiRD 2023 is an activity recognition competition using the OpenPack dataset. OpenPack dataset is a new large-scale multi-modal dataset of packing processes for human activity recognition. You'll develop a model to recognize the 10 work operations that make up packaging work in this challenge. The team that recognizes work operations with the highest accuracy will receive a cash prize and have their travel expenses covered to attend the awards ceremony at the Percom 2023 Workshop. Please visit our website at https://open-pack.github.io/challenge2022 for more detail.

Important Dates:

  • Competition Launch: 2022-10-15
  • Final Submission Deadline: 2023-01-TBA

Task

  • Recognize the 10 work operations in the packaging work.

Available Data

  • IMU (Acceleration, Gyroscope, Quaternion)
  • Depth Images, Keypoints,
  • etc

Web Site

Organizers

Takuya Maekawa (Osaka University)
Ken Yoda (Nagoya University)
Keisuke Fujii (Nagoya University)
Olga Heim (Doshisha University, present address: Leibniz institute for zoo and wildlife research)

Program committee

Computer science

Biology & neuroscience

Sponsors