Important Dates                          Paper Submissions                         Program                          Program Committee                          Speakers                          Organizers


Second Workshop on Affordances: Visual Perception of Affordances and Functional Visual Primitives for Scene Analysis (in conjunction with ECCV 2014), September 7, 2014, Zurich, Switzerland

The workshop seeks to address key challenges in computer vision and applications such as robotics with regard to functional form descriptions, which are termed as "affordances". Based on the Gibsonian principle of defining objects by their function, "affordances" have been studied extensively by psychologists and visual perception researchers, resulting in the creation of numerous cognitive models. These models are being increasingly revisited and adapted by computer vision researchers to build visual perception and behavioral algorithms in recent years. This workshop attempts to explore this nascent, yet rapidly emerging field of affordance based cognitive vision (recognition of objects, activities, scenes etc.) while integrating the efforts and language of affordance communities not just in computer vision, but also psychophysics and neurobiology by creating an open affordance research forum, feature framework and ontology called AfNet (theaffordances.net). In particular, the workshop will focus on emerging trends in affordances and other human-centered function/action features that can be used to build computer vision algorithms leading to various intelligent applications. The workshop will also feature contributions from researchers involved in traditional theories to affordances, especially from the point of view of psychophysics and neuro-biology. Avenues to aiding research in these fields using techniques from computer vision and cognitive robotics will also be explored. Primary topics addressed by the workshop include the following among others

Understanding various challenges in the field of affordances and building a common language and framework for communication across the varied affordance communities are the key goals of the proposed workshop. Through the course of the workshop, we also envisage the establishment of a working group for AfNet.

Location


ETHZ main building, Room D7.1

Poster Printing: For attendees wishing to print posters in Zurich, please note that there are no on-site facilities available. However, there are several print shops nearby.
Copyprint Main railway station (Shopville-Railcity), Mon-Sun: 07:45-20:00, (approx. 115 CHF)

Organizers

Karthik Mahesh Varadarajan Alireza Fathi Juergen Gall Markus Vincze

Program Committee

Program

Derek Hoiem Fei-Fei Li Abhinav Gupta Ashutosh Saxena Aaron Bobick

08:30 - 08:45 Introduction to Affordances - Karthik Mahesh Varadarajan - TU Wien

Session 1: Affordances in Computer Vision
08:45 - 09:30 Functional Indoor Scene Interpretation - Prof. Derek Hoiem - UIUC
09:30 - 09:35 CV1: Affordances in Video Surveillance (c) - Agheleh Yaghoobi, Hamed Rezazadegan Tavakoli*, Juha Roening - Univ. of Oulu       
09:35 - 09:50 CV2: Affordance-based Object Recognition using Interactions obtained from a Utility Maximization Principle (d) - Tobias Kluth, David Nakath*, Thomas Reineking, Christoph Zetzsche, Kerstin Schill - Univ. of Bremen       
09:50 - 10:05 CV3: Detecting Fine-grained Affordances with an Anthropomorphic Agent Model (d) - Viktor Seib*, Nicolai Wojke, Malte Knauf, Dietrich Paulus - Univ. of Koblenz-Landau       
10:05 - 10:35 Coffee Break + Posters
10:35 - 11:20 Let's Reason About Object Affordance - Prof. Fei-Fei Li - Stanford Univ.
11:20 - 12:00 Open discussion on papers and vision for future research
12:00 - 13:30 Lunch + Posters
13:30 - 14:15 Revisiting Gibsonian Affordances - Prof. Abhinav Gupta - CMU

Session 2: Psychophysics and Neurobiology/ Cognitive Modeling of Affordances
14:15 - 14:30 PN1: A Bio-Inspired Robot With Visual Perception of Affordances (d) - Oscar Chang* - Univ. Central de Venezuela         
14:30 - 14:45 PN2: Integrating Object Affordances with Artificial Visual Attention (d) - Jan Tuennermann*, Christian Born, Baerbel Mertsching - Univ. Paderborn       
14:45 - 15:00 PN3: Modeling Primate Control of Grasping for Robotic Applications (d) - Ashley Kleinhans* - CSIR S Africa, Serge Thill - Univ. of Skoevde, Benjamin Rosman - CSIR S Africa, Renaud Detry - Univ. of Liege, Bryan Tripp - Univ. of Waterloo         
15:00 - 15:20 Coffee Break + Posters
15:20 - 15:30 Posters QA Session

Session 3: Affordances in Cognitive Robotics
15:30 - 16:15 RoboBrain: Learning Physically-Grounded Affordances - Prof. Ashutosh Saxena - Cornell Univ.
16:15 - 16:30 CR1: OBEliSK: Novel Knowledgebase of Object Features and Exchange Strategies (d) - David Cabaneros Blanco*, Ana Belen Rodriguez Arias, Victor Fernandez-Carbajales Canete, Joaquin Canseco Suarez - Treelogic S.L.       
16:30 - 16:35 CR2: How Industrial Robots Benefit from Affordances (c) - Kai Zhou*, Martijn Rooker, Sharath Chandra Akkaladevi, Gerald Fritz, Andreas Pichler - Profactor GmbH         
16:35 - 16:50 CR3: Aspect Transition Graph: an Affordance-Based Model (d) - Li Yang Ku*, Shiraj Sen, Erik G. Learned-Miller, Roderic A. Grupen - Univ. of Mass. Amherst       
16:50 - 17:35 Affordances for Robotics* - Prof. Aaron Bobick - Georgia Tech.
17:35 - 18:15 Open discussion on papers and vision for future research
18:15 - 18:30 Closing Remarks

(c) - Conceptual Paper, (d) - Design Paper, (f) - Full Paper

Paper Submissions

Paper contributions to the workshop are solicited in four different formats. This departure from the regular format is intended to promote greater contribution and cater to the needs of affordance communities from various disciplines such as Knowledge Representation, Psychology, Psychophysics, Neuroscience, Kinematics, Ontologies besides traditional audience such as from Cognitive/ Computer Vision and Robotics.

Each contribution will be reviewed by three reviewers through a single-blind review process. The paper formatting should follow the ECCV formatting guidelines (Templates: LaTeX). All contributions are to be submitted via Microsoft Conference Management Tool in PDF. Please adhere to the following strict deadlines. In addition to direct acceptance, early submissions may be conditionally accepted, in which case submission of a revised version of the paper based on reviewer comments, prior to the late submission deadline is necessary. The final decision on acceptance of such conditionally accepted papers will be announced along with the decisions for the late submissions. Hence, while early submissions will have a better chance of acceptance than late submissions, submission to either (or both) rounds is equally encouraged. Also, note that the paper ID to be incorporated into the submission manuscript is available after performing an initial submission. If you still have issues finding your paper ID, it can be left blank for initial submissions, though it will be necessary to use the ID for the final ECCV version, OpenRsrch version, Springer version and ECCV attendance registration.

The program is composed of orals, posters as well as Pecha-Kucha style presentations. Papers, presentations and teaser slides will be archived on this website. A non-exclusive copyright grant is also transfered to ArXiv and OpenRsrch archiving systems that permit re-publishing alongwith an exclusive non-republishable copyright grant transfer to Springer Publishing.

Submission of Presentations and Posters

All accepted papers will be presented in the form of oral presentations as well as posters. Discussions over posters will be allocated to coffee and lunch breaks. Authors are expected to be present during the Posters QA Session. Please submit electronic versions of your posters in PDF for archiving via CMT. Printed posters are expected to be of A0 size. Oral presentations will be allocated 5 minutes (for conceptual papers), 15 minutes (for design/experimental papers) and 25 minutes (for full papers). A rough recommendation would be to have 1 slide per minute. Please also submit your presentations and posters (teaser slide) in PDF/PPT/PPTX format via CMT for achiving. Presentations will be loaded on a workshop laptop to avoid technical issues.

Springer Submission Guidelines

Authors of all accepted papers are required to submit Springer publication ready versions via the ECCV WS CMT. Note that this CMT site is different from the Affordances CMT. As the proceedings are published by Springer, the paper sources and also copyright forms have to be provided. Please refer to the Springer instructions on how to prepare the submission of the publication ready version:

Important Dates