Kihwan Kim, PhD Candidate

School of Interactive Computing, College of Computing, Georgia Institute of Technology

Date: Tuesday, December 6, 2011

Time: 1:00 pm – 3:00 pm EST

Location: Technology Square Research Building (TSRB) Room 223

Abstract

Analysis and visualization of dynamic scenes is often constrained by the amount of spatio-temporal information available from the environment. In most scenarios, we have to account for incomplete information and sparse motion data, requiring us to employ interpolation and approximation methods to fill for the missing information. Scattered data interpolation and approximation techniques have been widely used for solving the problem of completing surfaces and images with incomplete input data. We introduce approaches for such data interpolation and approximation from limited sensors, into the domain of analyzing and visualizing dynamic scenes. Data from dynamic scenes is subject to constraints due to the spatial layout of the scene and/or the configurations of video cameras in use. Such constraints include: (1) sparsely available cameras observing the scene, (2) limited field of view provided by the cameras in use, (3) incomplete motion at a specific moment, and (4) varying frame rates due to different exposures and resolutions.

In this thesis, we establish these forms of incompleteness in the scene, as spatio- temporal uncertainties, and propose solutions for resolving the uncertainties by applying scattered data approximation into a spatio-temporal domain.

The main contributions of this research are as follows: First, we provide an effi- cient framework to visualize large-scale dynamic scenes from distributed static videos. Second, we adopt Radial Basis Function (RBF) interpolation to the spatio-temporal domain to generate global motion tendency. The tendency, represented by a dense flow field, is used to optimally pan and tilt a video camera. Third, we propose a method to represent motion trajectories using stochastic vector fields. Gaussian Pro- cess Regression (GPR) is used to generate a dense vector field and the certainty of each vector in the field. The generated stochastic fields are used for recognizing motion patterns under varying frame-rate and incompleteness of the input videos. Fourth, we also show that the stochastic representation of vector field can also be used for modeling global tendency to detect the region of interests in dynamic scenes with camera motion. We evaluate and demonstrate our approaches in several applications for visualizing virtual cities, automating sports broadcasting, and recognizing traffic patterns in surveillance videos.

Committee:

• Prof. Irfan Essa (Advisor, School of Interactive Computing, Georgia Institute of Technology)
• Prof. James M. Rehg (School of Interactive Computing, Georgia Institute of Technology)
• Prof. Thad Starner (School of Interactive Computing, Georgia Institute of Technology)
• Prof. Greg Turk (School of Interactive Computing, Georgia Institute of Technology)
• Prof. Jessica K. Hodgins (Robotics Institute, Carnegie Mellon University, and Disney Research Pittsburgh)

Computational Journalism is defined as the application of computation to the activities of journalism such as information gathering, organization, communication, and dissemination of information, while upholding values of journalism such as accuracy and verifiability. Journalists are increasingly adopting and using the proliferation of open-source tools and embracing different styles of journalism. Explore how newsrooms are opening, what new tools are being created, and how to use those tools most effectively.

• Part of the Events associated with: FutureMedia Fest 2011 (Panel WebSite hosted by FutureMedia Fest 2011)
• Date and time: Tue, 11/15/2011 – 2:15 PM – 3:25 PM
• Location: Georgia Tech Hotel, Atlanta, GA 30308, USA.

Panelists:

• Irfan Essa, Professor, School of Interactive Computing, Georgia Tech [Bio] [Website] [Projects: CnJ].
• David Clinch, Clinch Media [Bio] [Project: Storyful.com to curatehttp://youtube.com/news and http://youtube.com/politics]
• Eric Gilbert, Assistant Professor, School of Interactive Computing, Georgia Tech [Bio] [Website] [Projects: comp.social].
• King-wa Fu, Research Assistant Professor, Journalism and Media Studies Centre University of Hong Kong. [Bio] [Project: Real-time Sina Weibo statistics
• John Perry, Database Specialist, Investigative Team, Atlanta Journal-Constitution.
• Leonard Witt, Robert D. Fowler Distinguished Chair in Communication at Kennesaw State University. [Bio] [Projects: JJIE.org, PJNet.org].

Topics of discussion will include (but will not be limited to):

• What is Computational Journalism?
• What impact has Computation / Information Technology / Networking Technology had on Journalism?
• What is the newsroom of the future? How has the newsroom changed?
• How has investigative journalism changed with new technologies?
• How is social networking changed how we gather, distribute, and share news (and information)?
• What are the economic / financial models that need to explored to support (and sustain) journalism?
• What is the role of an Editor in the new journalism model?
• What should we be teaching the next generation of journalists?

via CnJ Panel at Georgia Tech’s Future Media Fest 2011 | Computation + Journalism.

• Kim, Lee, and Essa (2011), “Gaussian Process Regression Flow for Analysis of Motion Trajectories,” in Proceedings of IEEE International Conference on Computer Vision (ICCV), 2011. [PDF] [WEBSITE] [VIDEO] [BIBTEX]

   @inproceedings{Kim2011-GPRF, Author = {K. Kim and D. Lee and I. Essa}, Booktitle = {Proceedings of IEEE International Conference on Computer Vision (ICCV)}, Month = {November}, Pdf = {http://www.cc.gatech.edu/~irfan/p/2011-Kim-GPRFAMT.pdf}, Publisher = {IEEE Computer Society}, Title = {Gaussian Process Regression Flow for Analysis of Motion Trajectories}, Url = {http://www.cc.gatech.edu/cpl/projects/gprf/}, Video = {http://www.youtube.com/watch?v=UtLr37hDQz0}, Year = {2011}}

Abstract

Analysis and Recognition of motions and activities of objects in videos requires effective representations for analysis and matching of motion trajectories. In this paper, we introduce a new representation specifically aimed at matching motion trajectories. We model a trajectory as a continuous dense flow field from a sparse set of vector sequences using Gaussian Process Regression. Furthermore, we introduce a random sampling strategy for learning stable classes of motions from limited data.

Our representation allows for incrementally predicting possible paths and detecting anomalous events from online trajectories. This representation also supports matching of complex motions with acceleration changes and pauses or stops within a trajectory. We use the proposed approach for classifying and predicting motion trajectories in traffic monitoring domains and test on several data sets. We show that our approach works well on various types of complete and incomplete trajectories from a variety of video data sets with different frame rates

http://www.youtube.com/watch?v=UtLr37hDQz0

via YouTube effects: Shake it like an Instagram picture — Online Video News.

YouTube users can now apply a number of Instagram-like effects to their videos, giving them a cartoonish or Lomo-like look with the click of a button. The effects are part of a new editing feature that also includes cropping and advanced image stabilization.

Taking the shaking out of video uploads should go a long way towards making some of the amateur footage captured on mobile phones more watchable, but it can also be resource-intensive — which is why Google’s engineers invented an entirely new approach toward image stabilization.

The new editing functionality will be part of YouTube’s video page, where a new “Edit video” button will offer access to filters and other editing functionality. This type of post-processing is separate from YouTube’s video editor, which allows to produce new videos based on existing clips.

http://www.youtube.com/watch?v=G-n9p28Yh8w

http://www.youtube.com/watch?v=i5keG1Y810U

This project, investigating formal languages as a general methodology for task transfer between distinct cyber-physical systems such as humans and robots, aims to expand the science of cyber physical systems by developing Motion Grammars that will enable task transfer between distinct systems.

Formal languages are tools for encoding, describing and transferring structured knowledge. In natural language, the latter process is called communication. Similarly, we will develop a formal language through which arbitrary cyber-physical systems communicate tasks via structured actions. This investigation of Motion Grammars will contribute to the science of human cognition and the engineering of cyber-physical algorithms. By observing human activities during manipulation we will develop a novel class of hybrid control algorithms based on linguistic representations of task execution. These algorithms will broaden the capabilities of man-made systems and provide the infrastructure for motion transfer between humans, robots and broader systems in a generic context. Furthermore, the representation in a rigorous grammatical context will enable formal verification and validation in future work.
Broader Impacts: The proposed research has direct applications to new solutions for manufacturing, medical treatments such as surgery, logistics and food processing. In turn, each of these areas has a significant impact on the efficiency and convenience of our daily lives. The PIs serve as coordinators of graduate/undergraduate programs and mentors to community schools. In order to guarantee that women and minorities have a significant role in the research, the PIs will annually invite K-12 students from Atlanta schools with primarily African American populations to the laboratories. One-day robot classes will be conducted that engage students in the excitement of hands-on science by interactively using lab equipment to transfer their manipulation skills to a robot arm.

Via Award#1146352 – EAGER: Linguistic Task Transfer for Humans and Cyber Systems.

Auto-Directed Video Stabilization with Robust L1 Optimal Camera Paths
Posted by Matthias Grundmann, Vivek Kwatra, and Irfan Essa,

Earlier this year, we announced the launch of new features on the YouTube Video Editor, including stabilization for shaky videos, with the ability to preview them in real-time. The core technology behind this feature is detailed in this paper, which will be presented at the IEEE International Conference on Computer Vision and Pattern Recognition (CVPR 2011).

Casually shot videos captured by handheld or mobile cameras suffer from significant amount of shake. Existing in-camera stabilization methods dampen high-frequency jitter but do not suppress low-frequency movements and bounces, such as those observed in videos captured by a walking person. On the other hand, most professionally shot videos usually consist of carefully designed camera configurations, using specialized equipment such as tripods or camera dollies, and employ ease-in and ease-out for transitions. Our goal was to devise a completely automatic method for converting casual shaky footage into more pleasant and professional looking videos.

http://www.youtube.com/watch?v=0MiY-PNy-GU

Our technique mimics the cinematographic principles outlined above by automatically determining the best camera path using a robust optimization technique. The original, shaky camera path is divided into a set of segments, each approximated by either a constant, linear or parabolic motion. Our optimization finds the best of all possible partitions using a computationally efficient and stable algorithm.

To achieve real-time performance on the web, we distribute the computation across multiple machines in the cloud. This enables us to provide users with a real-time preview and interactive control of the stabilized result. Above we provide a video demonstration of how to use this feature on the YouTube Editor. We will also demo this live at Google’s exhibition booth in CVPR 2011.

For more details see the Project Site. See the youtube video of the system on youtube. See the paper in PDF, and a technical video of the work.

Full paper is

• Matthias Grundmann, Vivek Kwatra, and Irfan Essa (2011), “Auto-Directed Video Stabilization with Robust L1 Optimal Camera Paths,” In Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition (CVPR 2011), Colorado Springs, CO, USA. [PDF][Video][Blog][Demo][Project Site]

• Grundmann, Kwatra, and Essa (2011), “Auto-Directed Video Stabilization with Robust L1 Optimal Camera Paths,” in Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2011.  [PDF] [WEBSITE][VIDEO] [DEMO][Google Research Blog] [BIBTEX]

   @inproceedings{2011-Grundmann-AVSWROCP, Author = {M. Grundmann and V. Kwatra and I. Essa}, Booktitle = {Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR)}, Month = {June}, Pdf = {http://www.cc.gatech.edu/~irfan/p/2011-Grundmann-AVSWROCP}, Publisher = {IEEE Computer Society}, Title = {Auto-Directed Video Stabilization with Robust L1 Optimal Camera Paths}, Url = {http://www.cc.gatech.edu/cpl/projects/videostabilization/}, Video = {http://www.youtube.com/watch?v=i5keG1Y810U}, Year = {2011}}

Abstract

We present a novel algorithm for automatically applying constrainable, L1-optimal camera paths to generate stabilized videos by removing undesired motions. Our goal is to compute camera paths that are composed of constant, linear and parabolic segments mimicking the camera motions employed by professional cinematographers. To this end, our algorithm is based on a linear programming framework to minimize the first, second, and third derivatives of the resulting camera path. Our method allows for video stabilization beyond the conventional filtering of camera paths that only suppresses high frequency jitter. We incorporate additional constraints on the path of the camera directly in our algorithm, allowing for stabilized and retargeted videos. Our approach accomplishes this without the need of user interaction or costly 3D reconstruction of the scene, and works as a post-process for videos from any camera or from an online source.

Abstract

My research group is focused on a variety of approaches for (a) low-level video analysis and synthesis and (b) recognizing activities in videos. In this talk, I will concentrate on two of our recent efforts. One effort aimed at robust spatio-temporal segmentation of video and another on using motion and flow to recognize and predict actions from video.

In the first part of the talk, I will present an efficient and scalable technique for spatio-temporal segmentation of long video sequences using a hierarchical graph-based algorithm. In this work, we begin by over segmenting a volumetric video graph into space-time regions grouped by appearance. We then construct a “region graph” over the obtained segmentation and iteratively repeat this process over multiple levels to create a tree of spatio-temporal segmentations. This hierarchical approach generates high quality segmentations, which are temporally coherent with stable region boundaries, and allows subsequent applications to choose from varying levels of granularity. We further improve segmentation quality by using dense optical flow to guide temporal connections in the initial graph. I will demonstrate a variety of examples of how this robust segmentation works, and will show additional examples of video-retargeting that use spatio-temporal saliency derived from this segmentation approach. (Matthias Grundmann, Vivek Kwatra, Mei Han, Irfan Essa, CVPR 2010, in collaboration with Google Research).

In the second part of this talk, I will show that constrained multi-agent events can be analyzed and even predicted from video. Such analysis requires estimating the global movements of all players in the scene at any time, and is needed for modeling and predicting how the multi-agent play evolves over time on the playing field. To this end, we propose a novel approach to detect the locations of where the play evolution will proceed, e.g. where interesting events will occur, by tracking player positions and movements over time. To achieve this, we extract the ground level sparse movement of players in each time-step, and then generate a dense motion field. Using this field we detect locations where the motion converges, implying positions towards which the play is evolving. I will show examples of how we have tested this approach for soccer, basketball and hockey. (Kihwan Kim, Matthias Grundmann, Ariel Shamir, Iain Matthews, Jessica Hodgins, Irfan Essa, CVPR 2010, in collaboration with Disney Research).

Time permitting, I will show some more videos of our recent work on video analysis and synthesis. For more information, papers, and videos, see my website.

via PhD Fellowships – Google Research.

Google PhD Fellowship Program Overview

Nurturing and maintaining strong relations with the academic community is a top priority at Google. The Google U.S./Canada PhD Student Fellowship Program was created to recognize outstanding graduate students doing exceptional work in computer science, related disciplines, or promising research areas. Last year we awarded 14 unique fellowships to some amazing students in the US and Canada:

• …
• Matthias Grundmann, Google U.S./Canada Fellowship in Computer Vision (Georgia Institute of Technology)
• …

Lights, Camera… EDIT! New Features for the YouTube Video Editor

Nine months ago we launched our cloud-based video editor. It was a simple product built to provide our users with simple editing tools. Although it didn’t have all the features available on paid desktop editing software, the idea was that the vast majority of people’s video editing needs are pretty basic and straight-forward and we could provide these features with a free editor available on the Web. Since launch, hundreds of thousands of videos have been published using the YouTube Video Editor and we’ve regularly pushed out new feature enhancements to the product, including:

• Video transitions (crossfade, wipe, slide)
• The ability to save projects across sessions
• Increased clips allowed in the editor from 6 to 17
• Video rotation (from portrait to landscape and vice versa – great for videos shot on mobile)
• Shape transitions (heart, star, diamond, and Jack-O-Lantern for Halloween)
• Audio mixing (AudioSwap track mixed with original audio)
• Effects (brightness/contrast, black & white)

A new user interface and project menu for multiple saved projects

While many of these are familiar features also available on desktop software, today, we’re excited to unveil two new features that the team has been working on over the last couple of months that take unique advantage of the cloud:

Stabilizer

Ever shoot a shaky video that’s so jittery, it’s actually hard to watch? Professional cinematographers use stabilization equipment such as tripods or camera dollies to keep their shots smooth and steady. Our team mimicked these cinematographic principles by automatically determining the best camera path for you through a unified optimization technique. In plain English, you can smooth some of those unsteady videos with the click of a button. We also wanted you to be able to preview these results in real-time, before publishing the finished product to the Web. We can do this by harnessing the power of the cloud by splitting the computation required for stabilizing the video into chunks and distributed them across different servers. This allows us to use the power of many machines in parallel, computing and streaming the stabilized results quickly into the preview. You can check out the paper we’re publishing entitled “Auto-Directed Video Stabilization with Robust L1 Optimal Camera Paths.” Want to see stabilizer in action? You can test it out for yourself, or check out these two videos. The first is without stabilizer.

http://www.youtube.com/watch?v=NnyYXbk8MXU

And now, with the stabilizer:

http://www.youtube.com/watch?v=sLpkdrjsRxY

An anthropomorphic robot arm and a human capture system enable the autonomous performance of assembly tasks with significant uncertainty in problem specifications and environments. This line of work is investigated through sequences of manipulation actions where the guarantee of the completion of task-level objectives is rooted in the discovery of the semantic structure of human manipulation. New research directions in anthropomorphic robotics are explored including programming by demonstration, activity recognition, control and estimation and planning.

The motion grammar laboratory infrastructure allows a great opportunity for research and education. New classroom experiences for undergraduates and graduates provide practical experience in robot human interaction and activity process sharing. This opens possibilities for human training and rehabilitation, as well as assistive personal robotic, and opens the door to a host of technological innovations.

via Award#1059362 – II-New: Motion Grammar Laboratory.

• Kim, Oh, Lee, and Essa (2011), “Augmenting aerial earth maps with dynamic information from videos,” Journal of Virtual Reality, Special Issue on Augmented Reality, vol. 15, iss. 2-3, pp. 1359-4338, 2011.  [PDF] [WEBSITE] [VIDEO] [DOI] [SpringerLink][BIBTEX]

  
  @article{2011-Kim-AAEMWDIFV,
   Author = {K. Kim and S. Oh and J. Lee and I. Essa},
   Doi = {10.1007/s10055-010-0186-2},
   Journal = {Journal of Virtual Reality, Special Issue on Augmented Reality},
   Number = {2-3},
   Pages = {1359-4338},
   Pdf = {http://www.cc.gatech.edu/~irfan/p/2011-Kim-AAEMWDIFV.pdf},
   Title = {Augmenting aerial earth maps with dynamic information from videos},
   Url = {http://www.cc.gatech.edu/cpl/projects/augearth},
   Video = {http://www.youtube.com/watch?v=TPk88soc2qw},
   Volume = {15},
   Year = {2011}}

Abstract

We introduce methods for augmenting aerial visualizations of Earth (from tools such as Google Earth or Microsoft Virtual Earth) with dynamic information obtained from videos. Our goal is to make Augmented Earth Maps that visualize plausible live views of dynamic scenes in a city. We propose different approaches to analyze videos of pedestrians and cars in real situations, under differing conditions to extract dynamic information. Then, we augment an Aerial Earth Maps (AEMs) with the extracted live and dynamic content. We also analyze natural phenomenon (skies, clouds) and project information from these to the AEMs to add to the visual reality. Our primary contributions are: (1) Analyzing videos with different viewpoints, coverage, and overlaps to extract relevant information about view geometry and movements, with limited user input. (2) Projecting this information appropriately to the viewpoint of the AEMs and modeling the dynamics in the scene from observations to allow inference (in case of missing data) and synthesis. We demonstrate this over a variety of camera configurations and conditions. (3) The modeled information from videos is registered to the AEMs to render appropriate movements and related dynamics. We demonstrate this with traffic flow, people movements, and cloud motions. All of these approaches are brought together as a prototype system for a real-time visualization of a city that is alive and engaging.

• Sarin, Kim, Essa, and Cooper (2011), “3-Dimensional Visualization of the Operating Room Using Advanced Motion Capture: A Novel Paradigm to Expand Simulation-Based Surgical Education,” in Proccedings of Society of Thoracic Surgeons Annual Meeting, Society of Thoracic Surgeons, 2011.  [BLOG][BIBTEX]

  
  @incollection{2011-Sarin-3VORUAMCNPESSE,
    Author = {E. L. Sarin and K. Kim and I. Essa and W. A. Cooper},
    Blog = {http://prof.irfanessa.com/2011/02/02/sts-2011/},
    Booktitle = {Proccedings of Society of Thoracic Surgeons Annual Meeting},
    Month = {January},
    Publisher = {Society of Thoracic Surgeons},
    Title = {3-Dimensional Visualization of the Operating Room Using Advanced Motion Capture: A Novel Paradigm to Expand Simulation-Based Surgical Education},
    Type = {Poster and Video Presentation},
    Year = {2011}}

A collaborative project between School of Interactive Computing, Georgia Institute of Technology, Atlanta, Georgia, Division of Cardiothoracic Surgery, Emory University School of Medicine, Atlanta, Georgia, and Inova Heart and Vascular Institute1, Fairfax, Virginia. This was a Video and a Poster presentation at the Society of Thoracic Surgeons Annual Meeting in San Diego, CA, Jan 2011.

Pei Yin, A. Criminisi, J. Winn, I. Essa (2011), “Bilayer Segmentation of Webcam Videos Using Tree-Based Classifiers” in Pattern Analysis and Machine Intelligence, IEEE Transactions on, Jan. 2011, Volume :  33 ,  Issue:1, ISSN :  0162-8828, Digital Object Identifier :  10.1109/TPAMI.2010.65,  IEEE Computer Society [Project Page|DOI]

ABSTRACT

This paper presents an automatic segmentation algorithm for video frames captured by a (monocular) webcam that closely approximates depth segmentation from a stereo camera. The frames are segmented into foreground and background layers that comprise a subject (participant) and other objects and individuals. The algorithm produces correct segmentations even in the presence of large background motion with a nearly stationary foreground. This research makes three key contributions: First, we introduce a novel motion representation, referred to as “motons,” inspired by research in object recognition. Second, we propose estimating the segmentation likelihood from the spatial context of motion. The estimation is efficiently learned by random forests. Third, we introduce a general taxonomy of tree-based classifiers that facilitates both theoretical and experimental comparisons of several known classification algorithms and generates new ones. In our bilayer segmentation algorithm, diverse visual cues such as motion, motion context, color, contrast, and spatial priors are fused by means of a conditional random field (CRF) model. Segmentation is then achieved by binary min-cut. Experiments on many sequences of our videochat application demonstrate that our algorithm, which requires no initialization, is effective in a variety of scenes, and the segmentation results are comparable to those obtained by stereo systems.

via IEEE Xplore – Abstract Page.

Kitware has received a $13,883,314 contract from Defense Advanced Research Projects Agency (DARPA) to develop a software system capable of automatically and interactively discovering actionable intelligence from wide area motion imagery (WAMI) of complex urban, suburban, and rural environments.

The primary information elements in WAMI data are moving entities in the context of roads, buildings, and other scene features. These entities, while exploitable, often yield fragmented tracks in complex urban environments due to occlusions, stops, and other factors. Kitware’s software system will use algorithmic solutions to associate tracks and then identify and integrate local events to detect potential threats and perform forensic analysis.

The developed algorithms will form the basis of a software prototype called the Persistent Stare Exploitation and Analysis System (PerSEAS) that will significantly augment an end-user’s ability to discover novel intelligence using models of activities, normalcy, and context. Since the vast majority of events are normal and pose no threat, the models must cross-integrate singular events to discover relationships and anomalies that are indicative of suspicious behavior or match previously learned – or defined – threat activity.

The advanced PerSEAS system will markedly improve an analyst’s ability to handle burgeoning WAMI data and reduce the time required to perform many current exploitation tasks, greatly enhancing the military’s capability to analyze and utilize the data for forensic analysis and through the issuance of timely threat alerts with a minimal number of false alarms.

Due to the complex, multi-disciplinary nature of the research, Kitware will partner with academic experts in the fields of computer vision, probabilistic reasoning, machine learning and other related domains. Phase I of the research is expected to be completed in two years.

The awarded contract will expand Kitware’s leadership in the field of computer vision, video analysis and advanced visualization software. The project will build upon our previous DARPA-sponsored research into content-based video retrieval on the VIRAT program; anomaly detection on the PANDA program; and the recognition of complex multi-agent activities in video.

To meet the PerSEAS program’s needs, Kitware has assembled a world-class team including four leading defense technology companies, Northrop Grumman Corporation, ; Honeywell Automation and Control Solutions Laboratories, Aptima, Inc., and Navia, Inc. As well as multiple internationally-renowned research institutions, including: the University of California, Berkeley; Computer Vision Laboratory, University of Maryland; Rensselaer Polytechnic Institute; the Computer Vision Lab at the University of Central Florida; the School of Interactive Computing at Georgia Tech and its affiliated Center for Robotics & Intelligent Machines; and Columbia University.

Abstract

Videos of multi-player team sports provide a challenging domain for dynamic scene analysis. Player actions and interactions are complex as they are driven by many factors, such as the short-term goals of the individual player, the overall team strategy, the rules of the sport, and the current context of the game. We show that constrained multi-agent events can be analyzed and even predicted from video. Such analysis requires estimating the global movements of all players in the scene at any time, and is needed for modeling and predicting how the multi-agent play evolves over time on the field. To this end, we propose a novel approach to detect the locations of where the play evolution will proceed, e.g. where interesting events will occur, by tracking player positions and movements over time. We start by extracting the ground level sparse movement of players in each time-step, and then generate a dense motion field. Using this field we detect locations where the motion converges, implying positions towards which the play is evolving. We evaluate our approach by analyzing videos of a variety of complex soccer plays.

• Matthias Grundmann, Vivek Kwatra, Mei Han, Irfan Essa (2010) “Discontinuous Seam-Carving for Video Retargeting” in Proceedings of IEEE Computer Vision and Pattern Recognition Conference (CVPR), San Francisco, CA, USA, June 2010 [PDF][Website][DOI][Video (Youtube)].

Abstract

We introduce a new algorithm for video retargeting that uses discontinuous seam-carving in both space and time for resizing videos. Our algorithm relies on a novel appearance-based temporal coherence formulation that allows for frame-by-frame processing and results in temporally discontinuous seams, as opposed to geometrically smooth and continuous seams. This formulation optimizes the difference in appearance of the resultant retargeted frame to the optimal temporally coherent one, and allows for carving around fast moving salient regions.

Additionally, we generalize the idea of appearance-based coherence to the spatial domain by introducing piece-wise spatial seams. Our spatial coherence measure minimizes the change in gradients during retargeting, which preserves spatial detail better than minimization of color difference alone. We also show that per-frame saliency (gradient- based or feature-based) does not always produce desirable retargeting results and propose a novel automatically computed measure of spatio-temporal saliency. As needed, a user may also augment the saliency by interactive region-brushing. Our retargeting algorithm processes the video sequentially, making it conducive for streaming applications.

Abstract

We present an efficient and scalable technique for spatio- temporal segmentation of long video sequences using a hierarchical graph-based algorithm. We begin by over- segmenting a volumetric video graph into space-time regions grouped by appearance. We then construct a “region graph” over the obtained segmentation and iteratively repeat this process over multiple levels to create a tree of spatio-temporal segmentations. This hierarchical approach generates high quality segmentations, which are temporally coherent with stable region boundaries, and allows subse- quent applications to choose from varying levels of granularity. We further improve segmentation quality by using dense optical flow to guide temporal connections in the initial graph.

We also propose two novel approaches to improve the scalability of our technique: (a) a parallel out- of-core algorithm that can process volumes much larger than an in-core algorithm, and (b) a clip-based process- ing algorithm that divides the video into overlapping clips in time, and segments them successively while enforcing consistency.

We demonstrate hierarchical segmentations on video shots as long as 40 seconds, and even support a streaming mode for arbitrarily long videos, albeit without the ability to process them hierarchically.

Abstract

We present a novel approach for robust localization of multiple people observed using multiple cameras. We usethis location information to generate sports visualizations,which include displaying a virtual offside line in soccer games, and showing players’ positions and motion patterns.Our main contribution is the modeling and analysis for the problem of fusing corresponding players’ positional informationas finding minimum weight K-length cycles in complete K-partite graphs. To this end, we use a dynamic programmingbased approach that varies over a continuum of being maximally to minimally greedy in terms of the numberof paths explored at each iteration. We present an end-to-end sports visualization framework that employs our proposed algorithm-class. We demonstrate the robustness of our framework by testing it on 60; 000 frames of soccerfootage captured over 5 different illumination conditions, play types, and team attire.

Part of the team led by Kitware Inc to work on Defense Advanced Research Projects Agency – Persistent Stare Exploitation and Analysis System (PerSEAS).

The Persistent Stare Exploitation and Analysis System (PerSEAS) program is developing the capability to automatically and interactively identify potential threats as they emerge based on the correlation of multiple disparate activities and events in wide area motion imagery (WAMI) and multi-INT data.  PerSEAS will enable new methods of threat hypothesis adjudication and forensic analysis through activity-based modeling and inferencing capabilities.

See

• DARPA BAA
• Wired Article on this project