RoF: Robust Features for Computer Vision

in conjunction with CVPR2016

Las Vegas, Nevada, June 26, 2016

Program (NEW)

See the end of this page.


1.      Paper submission is now open (Double blind review).

2.      The authors will submit full length papers (CVPR format) on-line, including (1) Title of paper & short abstract summarizing the main contribution, (2) Contributions must be written and presented in English, and (3) The paper in PDF format. All submissions will be peer-reviewed by at least 3 members of the program committee.



We encourage researchers to develop new robust features (e.g., manually designed local features or feature learning) to extract useful feature representations. We also encourage new theories and processes related to features. We are soliciting original contributions that address a wide range of theoretical and practical issues including, but not limited to:

  1. New features (manually designed local features, or feature learning in supervised, weakly supervised or unsupervised way) robust to noise, illuminations, scale, rotations and occlusions,
  2. Robust features that are suitable for wearable devices (e.g., smart glasses, smart phones) with strict requirements for computational efficiency and low power consumption,
  3. New applications of descriptors in different domains, e.g. medical domain,
  4. Other application in different  domain, such as one dimension (1D) digital signal processing, 2D images, 3D videos and 4D videos,
  5. Evaluations of current traditional descriptors,
  6. Evaluations between the features learned by deep learning and the traditional descriptors (e.g., LBP, SIFT, HOG)



The goal of the RoF Workshop 2016 is to accelerate the study of robustness of local descriptors in computer vision problems. With the increase of acceleration of digital photography and the advances in storage devices over the last decade, we have seen explosive growth in the available amount of visual data and equally explosive growth in the computational capacities for data understanding. How to extract robust representations for many computer vision tasks is still a challenging problem. This problem becomes more difficult when the data show different types of variations, e.g., noise, illuminations, scale, rotations and occlusions.


Important Dates

  1. Paper Submission: March 8, 2016, March 22, 2016
  2. Notification of acceptance: April 20, 2016
  3. Camera-ready paper: May 2nd, 2016
  4. Workshop (half day): July 1, 2016


Workshop Chairs


Invited Speakers:

1)      Deep features for face recognition and related problems

Prof. Rama Chellappa, University of Maryland, USA

This talk will cover their recent work on using deep features for face verification, recognition, age estimation, face detection, alignment and expression recognition.


2)      Convolutional patch representations for image retrieval

Dr. Cordelia Schmid, INRIA, FRANCE


Program Committee:

·           Aleix Martinez, Ohio State University, USA

·           Alice Caplier, Grenoble, France

·           Bin Fan, Chinese Academy of Sciences, China

·           Baochang Zhang Beihang University, China

·           Engin Tola, Aurvis R&D, Turkey

·           Enrique Alegre, University of León, Spain

·           Francesca Odone, university of Genova, Italy

·           Giovanni Fusco, Smith-Kettlewell Eye Research Institute, USA

·           Huu Tuan NGUYEN, Grenoble, France

·           Hazim Kemal Ekenel, Karlsruhe Institute of Technology, Germany

·           Ioannis Patras Queen Mary University, UK

·           Jean-Luc, Dugelay, Eurecom, France

·           Juho Kannala, University of Oulu, Finland

·           Jun Yang, Northwestern Polytechnical University, China

·           Lijun Yin, Binghamton University, USA

·           Lei Zhang, Hong Kong Polytechnic University,  Hong Kong, China

·           Loris Nanni, University of Padua (Padova), Italy

·           Michael Teutsch, Fraunhofer IOSB, Germany

·           Motilal Agrawal, Menlo Park, CA, USA

·           Nicoletta Noceti, University of Genova, Italy

·           Rainer Lienhart, Universität Augsburg, Germany

·           Ruiping Wang, Chinese Academy of Sciences, China

·           Rocio A Lizarraga-Morales, Universidad de Guanajuato DICIS, Mexico

·           Sei-ichiro Kamata, Waseda University, Japan

·           Shu Liao, Siemens, USA

·           Shengcai Liao, NLPR, Chinese Academy of Sciences, China

·           Tiago de Freitas Pereira, University of Campinas (UNICAMP), Brazil

·           Tri Huynh, Eurecom, France

·           Wenchao Zhang, Nanyang technological university, Singapore

·           Xianbiao Qi, Beijing University of Posts and Telecommunications, China

·           Xiaoyang Tan, Nanjing University of Aeronautics and Astronautics, China

·           Xiujuan Chai, ICT, Chinese Academy of Sciences, China

·           Xiaopeng Hong, University of Oulu, Finland




Jie Chen


Center for Machine Vision Research (CMV),

University of Oulu, Finland


(The following program was submitted to CVPR. please refer to CVPR program as the final schedule)




Paper/Talk Title






Oral session 1






Invited Talk1: Convolutional patch representations for image retrieval

 Dr. Cordelia Schmid (INRIA, France)



Fast Image Gradients Using Binary Feature Convolutions

Pierre-Luc St-Charles, Guillaume-Alexandre Bilodeau, Robert Bergevin



Texture Complexity based Redundant Regions Ranking for Object Proposal

Wei Ke, Tianliang Zhang, Jie Chen, Fang Wan, Qixiang Ye



Deeply Exploit Depth Information for Object Detection

Saihui Hou





Poster session


Poster session - including coffee




Efficient Deep Feature Learning and Extraction via StochasticNets

Mohammad Javad Shafiee, Parthipan Siva, Paul Fieguth, Alexander Wong



Embedding Sequential Information into Spatiotemporal Features for Action Recognition

Yuancheng Ye, YingLi Tian



Learning Discriminative Features with Class Encoder

Hailin Shi



Do We Need Binary Features for 3D Reconstruction?

Bin Fan, Qingqun Kong, Wei Sui, Zhiheng Wang, Xinchao Wang, SHIMING XIANG, Chunhong Pan, Pascal Fua



Deep Features or Not: Temperature and Time Prediction in Outdoor Scenes

Anna Volokitin, Radu Timofte, Luc Van Gool



Euclidean and Hamming Embedding for Image Patch Description with Convolutional Networks

Zishun Liu, Zhenxi Li, Juyong Zhang, Ligang Liu



Robust 2DPCA and Its Application

Quanxue Gao, Qianqian Wang



Background Subtraction using Local SVD Binary Pattern feature

Li Li Guo, Xin Liu, Zhen Ping Qiang, Dan Xu



Generating Discriminative Object Proposals via Submodular Ranking

Zhang Yangmuzi, Zhuolin Jiang, Xi Chen, Larry Davis



Time series representation and similarity based on local autopatterns

Mustafa Gökçe Baydoğan, George Runger





Oral session 2






Keynote Talk 2: Deep features for face recognition and related problems

Prof. Rama Chellappa (University of Maryland, USA)



Improving Gradient Histogram Based Descriptors for Pedestrian Detection in Datasets with Large Variations

Prashanth Balasubramanian, Sarthak Pathak, Anurag Mittal



Unsupervised Robust Feature-based Partition Ensembling to Discover Categories

Roberto Javier López-Sastre



The Best of Both Worlds: Combining Data-independent and Data-driven Approaches for Action Recognition

Zhenzhong Lan, Shoou-I Yu, Dezhong Yao, Ming Lin, Bhiksha Raj, Alexander Hauptmann



concluding remarks




Lunch (on your own)






Any question about this page? Turn to its host!

This page is maintained currently by jie chen. Last modified: 2016-1-21, 11:47