BMVA Workshop on Context Aware Cognitive Systems

Friday 17th of July 2015, @ the British Computer Society (BCS) in London (link to registration)

Chairs: Nicolas Pugeault, Sinan Kalkan, Frank Guerin and Angelo Cangelosi

Context is the totality of the information characterising the situation of a cognitive system; e.g. it can include objects, persons, places, and temporally extended information related to ongoing tasks, but also information not directly related to these tasks. The way natural cognitive systems respond to a given stimulus also depends on other, apparently irrelevant, stimuli which constitute what we call context. In fact natural systems not only consider these “apparently irrelevant” stimuli, they use them to their advantage: there is a wealth of evidence that context is used routinely for reasoning and disambiguation, for tasks as varied as object recognition (eg., Palmer, 1975; Torralba, 2010), categorisation (Bar, 2004), language understanding (Coventry et al., 2010), problem solving (Cheng & Holyoak, 1985).

In contrast, artificial cognitive systems have long considered context as spurious, to be ignored, factored out or minimised in lab environments. This is now changing: in the recent years, there has been increasing efforts from the robotics and AI communities to research and design systems that can perform outside of closely controlled lab environments into complex, confusing, real-world situations. This will require a new generation of artificial systems, endowed with the cognitive tools to detect, assess, process and learn contextual relations (eg., Choi et al 2012; Li et al, 2012). This problem is intrinsically cross-disciplinary, touching aspects of computer vision, robotics, artificial intelligence, machine learning and psychology.

Keynotes

Important Dates

Sponsors

This meeting is sponsored by the British Machine Vision Association (BMVA) and the Poeticon++ project.

Call for papers

This one day workshop will be the first interdisciplinary colloquium to discuss the multiple facets of context and the challenges it poses to cognitive systems. Contributions are welcome from all related fields, including new experimental results and models as well as position or review papers. Topics of interest include, but are not limited to:

The workshop will consist in a mixture of oral presentations, one poster session and a panel discussion.

Submissions format

Contributors should submit an extended abstract of 2 pages via the submission website. Submissions will be peer reviewed and oral and selected contributions will be allocated oral or poster presentations.

+++ link to the submission website +++

References

Bar, M. (2004) “Visual objects in context.” Nature Reviews Neuroscience 5(8):617–629.

Cheng, P. W., & Holyoak, K. J. (1985). “Pragmatic reasoning schemas.” Cognitive Psychology,17:391-416.

Congcong Li, Adarsh Kowdle, Ashutosh Saxena, Tsuhan Chen (2012) “Towards Holistic Scene Understanding: Feedback Enabled Cascaded Classification Models.” IEEE Trans Pattern Analysis and Machine Intelligence (PAMI), 34(7):1394-1408.

Coventry, KR, Lynott, D, Cangelosi, A, Monrouxe, L, Joyce, D and Richardson, DC (2010) “Spatial language, visual attention, and perceptual simulation.” Brain and Language 112(3):202-213.

Myung J.C., Torralba, A. and Willsky, A.S., (2012) "A Tree-Based Context Model for Object Recognition." IEEE Trans Pattern Analysis and Machine Intelligence (PAMI), 34(2):240-252.

Palmer SE (1975) “The effects of contextual scenes on the identification of objects.” Memory & Cognition, 3(5):519-26.

Torralba, A. (2003) “Contextual priming for object detection.” International Journal of Computer Vision, 53(2):169-191.