<div dir="ltr">==============================<wbr>========================<br>
## 2nd Call for Papers ##<br>
==============================<wbr>========================<br>
ICCV workshop on Transferring and Adapting Source Knowledge in Computer Vision (TASK-CV) 2017<br>
Venice, Italy, 29 October, 2017<br>
<br>
Workshop site: <a href="http://adas.cvc.uab.es/task-cv2017/" rel="noreferrer" target="_blank">http://adas.cvc.uab.es/task-<wbr>cv2017/</a><br>
Challenge site: <a href="http://ai.bu.edu/visda-2017/" rel="noreferrer" target="_blank">http://ai.bu.edu/visda-2017/</a><br>
<br>
************************<br>
Important Dates<br>
************************<br>
Challenge devkit/data release: see the challenge website!<br>
<br>
Submission deadline: June 21st, 2017<br>
Author notification: August 4th, 2017<br>
Camera-ready due: August 14th, 2017<br>
Workshop date: 29 October, 2017<br>
<br>
************************<br>
Scope<br>
************************<br>
<br>
This would be the 4th annual workshop that brings together computer
vision researchers interested in domain adaptation and knowledge
transfer techniques. New this year would be the proposed Domain
Adaptation Challenge, see below.<br>
<br>
A key ingredient of the recent successes in computer vision has been the
availability of visual data with annotations, both for training and
testing, and well-established protocols for evaluating the results.
However, this traditional supervised learning framework is limited when
it comes to deployment on new tasks and/or operating in new domains. In
order to scale to such situations, we must find mechanisms to reuse the
available annotations or the models learned from them.<br>
<br>
************************<br>
Topics<br>
************************<br>
<br>
Accordingly, TASK-CV aims to bring together research in transfer
learning and domain adaptation for computer vision and invites the
submission of research contributions on the following topics:<br>
<br>
-TL/DA learning methods for challenging paradigms like unsupervised, and incremental or on-line learning.<br>
-TL/DA focusing on specific visual features, models or learning algorithms.<br>
-TL/DA jointly applied with other learning paradigms such as reinforcement learning.<br>
-TL/DA in the era of deep neural networks (e.g., CNNs), adaptation
effects of fine-tuning, regularization techniques, transfer of
architectures and weights, etc.<br>
-TL/DA focusing on specific computer vision tasks (e.g., image
classification, object detection, semantic segmentation, recognition,
retrieval, tracking, etc.) and applications (biomedical, robotics,
multimedia, autonomous driving, etc.).<br>
-Comparative studies of different TL/DA methods.<br>
-Working frameworks with appropriate CV-oriented datasets and evaluation protocols to assess TL/DA methods.<br>
-Transferring knowledge across modalities (e.g., learning from 3D data
for recognizing 2D data, and heterogeneous transfer learning)<br>
-Transferring part representations between categories.<br>
-Transferring tasks to new domains.<br>
-Solving domain shift due to sensor differences (e.g., low-vs-high
resolution, power spectrum sensitivity) and compression schemes.<br>
-Datasets and protocols for evaluating TL/DA methods.<br>
<br>
This is not a closed list; thus, we welcome other interesting and relevant research for TASK-CV.<br>
<br>
******************************<wbr>*******<br>
Domain Adaptation Challenge<br>
******************************<wbr>*******<br>
<br>
This year we are pleased to announce an accompanying Visual Domain Adaptation challenge. Please see the challenge website ( <a href="http://ai.bu.edu/visda-2017/" rel="noreferrer" target="_blank">http://ai.bu.edu/visda-2017/</a> ) for details, dates, and submission guidelines.<br>
<br>
************************<br>
Best Paper Award<br>
************************<br>
<br>
As in previous workshops, we plan to award 1-2 best papers. Sponsors will be announced.<br>
<br>
************************<br>
Submission<br>
************************<br>
- All submissions will be handled via the CMT website <a href="https://cmt3.research.microsoft.com/TASKCV2017/" rel="noreferrer" target="_blank">https://cmt3.research.<wbr>microsoft.com/TASKCV2017/</a>.<br>
- The format of the papers is the same as the ICCV main conference. The
contributions will consist in Extended Abstracts (EA) of 6 pages
(excluding the references).<br>
- We accept dual submissions to ICCV 2016 and TASK-CV 2017. In other
words, the submission to TASK-CV 2017 should be a 6-page summary of the
submission to ICCV 2017 (Authors need to indicate the difference in
their ICCV camera-ready version after their paper accepted by ICCV).<br>
- Submissions will be rejected without review if they: exceed the page limitation or violate the double-blind policy.<br>
- Manuscript templates can be found at the main conference website: <a href="http://iccv2017.thecvf.com/submission/main_conference/author_guidelines" rel="noreferrer" target="_blank">http://iccv2017.thecvf.com/<wbr>submission/main_conference/<wbr>author_guidelines</a><br>
- The accepted papers will be included in the ICCV workshop collections, and also linked in the TASK-CV webpage.<br>
<br>
<br>
************************<br>
Workshop Chairs<br>
************************<br>
<br>
Tatiana Tommasi, University of Rome La Sapienza, Italy<br>
Kate Saenko, Boston University, USA<br>
Ben Usman, Boston University, USA<br>
Xingchao Peng, Boston University, USA<br>
Judy Hoffman, Stanford, USA<br>
Dequan Wang, UC Berkeley, USA<br>
Antonio M. López, Computer Vision Center & U. Autònoma de Barcelona, Spain<br>
Wen Li – ETH Zurich, Switzerland<br>
Francesco Orabona, Stony Brook University, USA<br>
David Vázquez, Computer Vision Center & U. Autònoma de Barcelona, Spain</div>