International Workshop on Large Scale Visual Recognition and Retrieval

6/12/15, Boston, MA, CVPR 2015

Room 102



 9:00    Introduction

Places Database: Large Scale Visual Scene Recognition 
Antonio Torralba
Massachusetts Institute of Technology
 10:00  Coffee break


Learning Deep Structured Models

Raquel Urtasun
, University of Toronto

DeepDive: Opportunities for Dark Data in Text and Images
Christopher Ré, Stanford University
 12:00   Poster spotlights
 12:30   Lunch & Posters


Scaling Up and Diversifying the Visual Knowledge
Abhinav Gupta, Carnegie Mellon University
14:45    Awards

Bigger than Bigger: Very Large-scale Scene Understanding    
Jianxiong Xiao
, Princeto
n University
 15:35  Coffee break


Scalable approaches for large scale vision

Christian SzegedyGoogle


Poster spotlights are at noon. Poster session is 12:30-14:00.

  • [Best poster award] An active search strategy for efficient object class detection. Abel Gonzalez-Garcia, Alexander Vezhnevets and Vittorio Ferrari
  • SUN RGB-D: A RGB-D Scene Understanding Benchmark SuiteShuran Song, Samuel Lichtenberg, Jianxiong Xiao
  • Visual Census: Using Cars to Study People and Society. Timnit Gebru, Jonathan Krause, Yilun Wang, Jia Deng, Li Fei-Fei
  • Max-Margin, Single-Layer Adaptation of Transferred Image FeaturesPraveen Kulkarni, Joaquin Zepeda, Frederic Jurie, Patrick Perez, Louis Chevallier
  • PanoContext: A Whole-room 3D Context Model for Panoramic Scene UnderstandingYinda Zhang, Shuran Song, Ping Tan, Jianxiong Xiao
  • Linking Past to Present: Discovering Style in Two Centuries of Architecture. Stefan Lee, Nicolas Maisonneuve, David Crandall, Alyosha Efros, Josef Sivic
  • Every Moment Counts: A Dataset for Dense Detailed Labeling of Actions in Complex Videos. Serena Yeung, Olga Russakovsky, Ning Jin, Mykhaylo Andriluka, Greg Mori, Fei-Fei Li
  • Visual Search Experiments at Pinterest. Dmitry Kislyuk, David Liu, Kevin Jing, Andrew Zhai 
  • Hyper-class Augmented and Regularized Deep Learning for Fine-grained Image Classification. Saining Xie, Tianbao Yang, Xiaoyu Wang, Yuanqing Lin
  • Fine-Grained Visual Categorization via Multi-stage Metric Learning. Qi Qian, Rong Jin, Shenghuo Zhu, Yuanqing Lin
  • Joint calibration of Ensemble of Exemplar SVMs. Davide Modolo, Alexander Vezhnevets, Olga Russakovsky and Vittorio Ferrari

Call for submission

Big data has been revolutionizing computer vision research. The increasing availability of massive visual datasets creates unprecedented opportunities for researchers to tackle fundamental computer vision challenges: recognizing everything in the visual world, indexing and organizing the sea of visual information, and extracting knowledge and discovering patterns from big visual data. Achieving these goals calls for bold innovations on many fronts: data collection, learning, inference, representations, indexing, and systems infrastructure. 

The goal of this workshop is providing a venue for researchers interested in large-scale vision to present new work, exchange ideas, and connect with each other. The workshop will feature invited talks from leading researchers as well as a poster session that fosters in depth discussion. 

We invite submissions of extended abstracts related to the following topics in the context of big data and large-scale vision:

  • Indexing algorithms and data structures
  • Weakly supervised or unsupervised learning
  • Metric learning
  • Visual models and feature representations
  • Transfer learning and domain adaptation
  • Systems and infrastructure
  • Visual data mining and knowledge discovery
  • Dataset issues (e.g. dataset collection and dataset biases
  • Efficient learning and inference techniques
  • Optimization techniques
The abstracts should be no more than 2 pages in CVPR 2015 format. Accepted abstracts will be presented as posters. The workshop is not intended as a venue for publication and no proceedings will be produced. All submissions will undergo double-blind reviews. In the case of previous published work, the review will be single-blind.

Important Dates
  • Submission deadline: May 1, 2015
  • Notification date: May 22, 2015
  • Workshop date: June 12, 2015

Program Committee
  • Alexandre Alahi, Stanford University
  • Drago Anguelov, Google
  • Lamberto Ballan, Stanford University
  • Serge Belongie, Cornell
  • Alex Berg, UNC Chapel Hill
  • Tamara Berg, UNC Chapel Hill
  • Lubomir Bourdev, Facebook
  • Trevor Darrell, UC Berkeley
  • Alyosha Efros, UC Berkeley
  • Vittorio FerrariUniversity of Edinburgh
  • Sanja Fidler, University of Toronto
  • Ross Girshick, Microsoft Research
  • Kristen Grauman, UT Austin
  • Jonathan Krause, Stanford University
  • Christoph Lampert, IST Austria
  • Ce Liu, Microsoft Research
  • Greg Mori, Simon Fraser University
  • Florent Perronnin, XRCE
  • Fei Sha, USC
  • Karen SimonyanGoogle DeepMind
  • Noah Snavely, Cornell
  • Lorenzo Torresani, Dartmouth College
  • Xiaoyu Wang, Snapchat Research
  • Jianchao Yang, Adobe
  • Amir Zamir, Stanford University

Program Chairs
  • Olga Russakovsky, Stanford University (contact: olga - at -
  • Jason Corso, University of Michigan
  • Jia Deng, University of Michigan
  • Yuanqing Lin, NEC Labs America

General Chairs
  • Samy Bengio, Google
  • Fei-Fei Li, Stanford University