Mid-Atlantic Computer Vision (MACV) Workshop 2014

April 18, 2014
Brush Mountain A/B, Squires Student Center, Virginia Tech

https://www.flickr.com/photos/lordsutch/6238485771

Overview

Welcome to the first meeting of the Mid-Atlantic Computer Vision workshop. This first iteration of the workshop is organized by Devi Parikh and Dhruv Batra. We bring together computer vision researchers around the Mid-Atlantic region of the United States for an informal exchange of ideas in all areas of computer vision. We welcome people to present new research, work that will be presented at upcoming conferences, reviews of not-well-publicized work, and descriptions of "work in progress." We hope that these discussions will let everyone get feedback on their work, inspire new research directions, and maybe even kindle new collaborations.

Location

The workshop will be held in the Brush Mountain A/B rooms of the Squires Student Center at Virginia Tech. Squires is located at 290 College Avenue, Blacksburg, VA 24060.

Attendees

The list of attendees can be found here.

Schedule

Time	Event(s)
8:30 - 9:00	Light Breakfast and Name Tags at Squires
9:00 - 10:20 (15 min talk + 4 min questions + 1 min trans) x 4	"Sparse Subspace Clustering" Rene Vidal (JHU) "Jointly Learning Dictionaries and Subspace Structure for Video-based Face Recognition" Guangxiao Zhang (UMD), Ran He (UMD), Larry Davis (UMD) "Modes of a Discrete Graphical Model" Chao Chen (Rutgers) "Active Learning in Structured Probabilistic Models with Deterministic Samples" Qing Sun (VT), Ankit Laddha (VT), Faruk Ahmed (VT), Dhruv Batra (VT)
10:20 - 10:40	Break
10:40 - 12:20 (15 min talk + 4 min questions + 1 min trans) x 5	"Unsupervised Dense Object Discovery, Detection, Tracking, and Reconstruction" Lu Ma (GWU), Gabe Sibley (GWU) "Visual Interpretation of Manipulation Actions" Yezhou Yang (UMD), Ching Lik Teo (UMD), Cornelia Fermuller (UMD), Yiannis Aloimonos (UMD) "Predicting Gaze Behavior Using Gaze Fields" Hyun Soo Park (CMU), Eakta Jain (CMU), Yaser Sheikh (CMU) "Human Action Recognition by Representing 3D Human Skeleton as a Point in a Lie Group" Raviteja Vemulapalli (UMD), Felipe Arrate (UMD), Rama Chellappa (UMD) "Tracking Multiple People Online and in Real-Time" Ergys Ristani (Duke), Carlo Tomasi (Duke)
12:20 - 2:00	Lunch (on your own)
2:00 - 3:20 (15 min talk + 4 min questions + 1 min trans) x 4	"PatchMatch Based Joint View Selection and Depthmap Estimation" Enliang Zheng (UNC), Enrique Dunn (UNC), Vladimir Jojic (UNC), Jan-Michael Frahm (UNC) "Semantic Segmentation for Robot Perception" Jana Kosecka (GMU) "Forecast Plots: How Much Does Image Classiﬁcation Help Proposal-based Semantic Segmentation" Xiao Lin (VT), Michael Cogswell (VT), Devi Parikh (VT), Dhruv Batra (VT) "Thoracoabdominal Lymph-node Detection: Exploration of Shallow and Deep Models" Ari Self (NIH), Holger Roth (NIH), Le Lu (NIH), Jiamin Liu (NIH), Evrim Turkbey (NIH), Ronald Summers (NIH)
3:20 - 4:40	Break and Poster Session
4:40 - 6:20 (15 min talk + 4 min questions + 1 min trans) x 5	"Zero-Shot Learning via Visual Abstraction" Stanislaw Antol (VT), Larry Zitnick (MSR), Devi Parikh (VT) "How Hipster Are You?" Hadi Kiapour (UNC), Kota Yamaguchi (Stony Brook), Alexander Berg (UNC), Tamara Berg (UNC) "GREEN: A New Automated Evaluation Metric for Image to Text" Ramakrishna Vedantam (VT), Larry Zitnick (MSR), Devi Parikh (VT) "Joint Summarization of Large-scale Collections of Web Images and Videos for Storyline Reconstruction" Gunhee Kim (Disney Research), Leonid Sigal (Disney Research), Eric P. Xing (CMU) "Learning Connections between Language and Vision from Large Scale Data" Vicente Ordonez (UNC), Jia Deng (UM), Yejin Choi (Stony Brook), Alexander Berg (UNC), Tamara Berg (UNC)

Poster Program

Title	Authors
Features for 3D Hand Tracking	Cassandra Carley (Duke), Carlo Tomasi (Duke)
Introspective Semantic Segmentation	Gautam Singh (GMU), Jana Kosecka (GMU)
A Three-tiered Bottom-up Approach for Supervised Pancreas Segmentation: Improving the State-of-the-Art Methods	Amal Farag (NIH), Le Lu (NIH), Jiamin Liu (NIH), Evrim Turkbey (NIH), Ronald Summers (NIH)
Bottom-up 3D Bone Parsing: A PASCAL Representation	Le Lu (NIH), Dijia Wu (Microsoft), Nathan Lay (Siemens), Ronald Summers (NIH), Kevin S. Zhou (Siemens)
DCMSVM: Distributed Parallel Training For Single-Machine Multi-class Classifiers	Xufeng Han (UNC), Alexander Berg (UNC)
Runway to Realway: Visual Analysis of Fashion	Sirion Vittayakorn (UNC), Kota Yamaguchi (Stony Brook), Alexander Berg (UNC), Tamara Berg (UNC)
Streaming Training and Updating Object Detectors on Images and Videos	Wei Liu (UNC), Alexander Berg (UNC)
Computer Vision Threats for Security	Yi Xu (UNC), Fabian Monrose (UNC), Jan-Michael Frahm (UNC)
Comparative Evaluation of Binary Descriptors	Jared Heinly (UNC), Enrique Dunn (UNC), Jan-Michael Frahm (UNC)
Combining Semantic Scene Priors and Haze Removal for Single Image Depth Estimation	Ke Wang (UNC), Enrique Dunn (UNC), Joseph Tighe (UNC), Jan-Michael Frahm (UNC)
Incremental and Adaptive Front-End Vincent Fusion	Vincent Spinella-Mamo (GWU), Juan Falquez (GWU), Gabe Sibley (GWU)
Scalable Multi-device SLAM	Jack Morrison (GWU), Gabe Sibley (GWU)
Asynchronous Adaptive Conditioning for Visual-Inertial SLAM	Nima Keivan (GWU), Alonso Patron (GWU), Gabe Sibley (GWU)
Affordance of Object Parts from Geometric Features	Austin Myers (UMD), Angjoo Kanazawa (UMD), Cornelia Fermuller (UMD), Yiannis Aloimonos (UMD)
Ray Saliency: Bottom-Up Visual Saliency for a Rotating and Zooming Camera	Garrett Warnell (UMD), Philip David (UMD), Rama Chellappa (UMD)
Monte Carlo Acquired Directions for Domain Adaptation	David Shaw (UMD), Rama Chellappa (UMD)
Submodular Reranking with Multiple Feature Modalities	Fan Yang (UMD), Zhuolin Jiang (Huawei Technologies), Larry Davis (UMD)
Adding Unlabeled Samples to Categories by Learned Attributes	Jonghyun Choi (UMD), Mohammad Rastegari (UMD), Ali Farhadi (UW), Larry Davis (UMD)
Predictable Dual-View Hashing	Mohammad Rastegari (UMD), Jonghyun Choi (UMD), Hal Daumé (UMD), Larry Davis (UMD)
Dog Breed Classification Using Part Localization	Jiongxin Liu (Columbia), Angjoo Kanazawa (UMD), Peter Belhumeur (Columbia), David Jacobs (UMD)
Objecthood Prior-based Dynamic Scene Segmentation	Yezhou Yang (UMD), Cornelia Fermuller (UMD), Yiannis Aloimonos (UMD)
A Mid-Level Approach for Efficient Object Search in Clutter	Ching Teo (MUD), Cornelia Fermuller (UMD), Yiannis Aloimonos (UMD)
Shadow-free Segmentation in Still Images Using Local Density Measure	Aleksandrs Ecins (UMD), Cornelia Fermüller (UMD), Yiannis Aloimonos (UMD)
Computationally Efficient Regression on a Dependency Graph for Human Pose Estimation	Kota Hara (UMD), Rama Chellappa (UMD)
Generating Multiple Structured Visual Interpretations	Debadeepta Dey (CMU), Varun Ramakrishna (CMU), Martial Hebert (CMU), J. Andrew (Drew) Bagnell (CMU)
Data-driven Exemplar Model Selection	Ishan Misra (CMU)
Visual Chunking: A List Prediction Framework for Region-Based Object Detection	Nick Rhinehart (CMU), Jiaji Zhou (CMU), Martial Hebert (CMU), J. Andrew (Drew) Bagnell (CMU)
ActionReaction: Forecasting the Dynamics of Dual Agent Interactions	De-An Huang (CMU), Kris M. Kitani (CMU)
Assessment of Psychophysiological Characteristics of Drivers Using Heart Rate from naturalistic Face Video Data	Abhijit Sarkar (VT), A. Lynn Abbott (VT), Zachary Doerzaph (VTTI)
Discovering the Language of Surgery	Colin Lea (JHU)
Hierarchical Joint Max-Margin Learning of Mid- and Top-Level Representations for Visual Recognition	Hans Lobel (JHU), Rene Vidal (JHU), Alvaro Soto (Pontificia Universidad Católica de Chile)
Joint Action Classification and Segmentation Using a Markov/semi-Markov Conditional Random Field	Lingling Tao (JHU), Luca Zapella (JHU), Greg Hager (JHU), Rene Vidal (JHU)
Efficient Object Localization and Pose Estimation with 3D Wireframe Models	Erdem Yoruk (JHU), Rene Vidal (JHU)
Active Clustering and Subclustering	Arijit Biswas (UMD), David Jacobs (UMD)
Predicting Failures of Vision Systems	Peng Zhang (VT), Jiuling Wang (UTA), Ali Farhadi (UW), Martial Hebert (CMU), Devi Parikh (VT)
Holistic Scene Understanding using Diverse Solutions	Ankit Laddha (VT), Dhruv Batra (VT)
Estimating Masses around Solutions in Structured Output Spaces	Faruk Ahmed (VT), Danny Tarlow (MSR), Dhruv Batra (VT)
SegNets: Semantic Segmentation with Deep Neural Networks	Michael Cogswell (VT), Dhruv Batra (VT)
Predicting User Annoyance Using Visual Attributes	Gordon Christie (VT), Amar Parkash (Goibibo), Ujwal Krothapalli (VT), Devi Parikh (VT)
CloudCV: Large Scale Distributed Computer Vision as a Cloud Service	Neelima Chavali (VT), Harsh Agarwal (VT), Abdullah Alfadda (VT), Clint Solomon (VT), Prakriti Banik (VT), Dhruv Batra (VT)
Interactively Guiding Semi-Supervised Clustering via Attribute-based Explanations	Shrenik Lad (VT), Devi Parikh (VT)
Scene Parsing with Object Instances and Occlusion Ordering	Joseph Tighe (UNC), Marc Niethammer (UNC), Svetlana Lazebnik (UIUC)
Segmentation of Microscopic Images in Biology	Branka Lakic (Duke), Carlo Tomasi (Duke)

Posters

Please create your posters in the standard portrait style (i.e., 4 feet tall by 3 feet wide). There is limited space for posters, so please do not make your posters any wider.

Talks

Please keep presentations to 15 minutes in length.

Getting to Campus

By Bus

Blacksburg Transit (BT) provides a great way to get to campus without needing to deal with parking. The route map can be found here or you can use Google Maps to figure out directions and times. Live updates of the bus routes can be found via BT4U.

By Car

New Parking Information
There is an undergraduate admissions event on campus on April 18th. The Visitor Center will be very busy, and finding parking on campus may be harder than usual. If walking or taking the bus to campus is not possible, we recommend the following:

Public Parking (Recommended)
Kent Square Garage - $4, open 24 hours, and 6 minute walk to Squires

View Larger Map

On-campus Parking (Not Recommended)
Passes are available at the Visitor Center (925 Prices Fork Road, near The Inn at Virginia Tech). You will have to mention that you are attending the Mid-Atlantic Computer Vision workshop to get your pass. Please keep in mind that on-campus parking operates on a first-come basis, so no guarantee can be made about parking availability. An explanation of which parking spots you can occupy can be found here. A map of all parking locations can be found here. Following is a list of some parking lots that might be convenient:

in the Squires Lot (south of Squires)
on Alumni Mall (north of Squires)
around the Drillfield (west of Squires)
in Perry St Lot 3 (located at the corner of Perry St and Stanger St)
the Perry St Garage (north side of campus)

Campus Maps

A detailed map of campus can be found here.
An interactive map can be found here.

Wireless Internet on Campus

We have e-mailed everyone the procedure for getting wireless Internet on campus. Please try to set this up several days before the workshop to avoid technical difficulties.

Accommodation

We encourage you to make accommodation arrangements as soon as possible, since vacancies might fill up quickly due to visiting families. A few (not all) options are below:

Contact Information

If you have any further questions about the workshop or comments about the website, please contact the MACV Kingpin, Stan Antol, at santol-at-vt.edu (replacing -at- with @).