AJR Your Link to CME
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 QUICK SEARCH:   [advanced]


     


This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Tan, L. K.
Right arrow Articles by Ng, K.-H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tan, L. K.
Right arrow Articles by Ng, K.-H.
Social Bookmarking
 Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?
DOI:10.2214/AJR.05.0994
AJR 2006; 186:898-901
© American Roentgen Ray Society

A Low-Cost, High-Quality Method for Remote Recording of Presentations

Li Kuo Tan1, Jeannie Hsiu Ding Wong1 and Kwan-Hoong Ng1

1 All authors: Department of Radiology, Faculty of Medicine, University of Malaya, Kuala Lumpur, Malaysia 59100.

Received June 10, 2005; accepted after revision August 26, 2005.

 
Address correspondence to L. K. Tan.


Abstract
Top
Abstract
Introduction
Existing Methods
Materials and Methods
Discussion
References
 
OBJECTIVE. The purpose of this article was to develop a low-cost method for high-quality remote capturing and recording of multimedia presentations.

CONCLUSION. The method for making high-quality digital recordings of presentations by using the mix of freely available software presented in this article offers great potential for institutions on a budget as well as others seeking viable alternatives to current methods of presentation recording.

Keywords: computers • education


Introduction
Top
Abstract
Introduction
Existing Methods
Materials and Methods
Discussion
References
 
Many benefits exist in maintaining a digital archive of presentations. Routine academic lectures, case reports, conference presentations, and so on can be easily retrieved for viewing by students and doctors alike. In addition, many authors have discussed the advantages of digital streaming [1], enabling a far wider audience to be reached over the Internet.

In the past decade, there has been a notable trend away from traditional presentation tools such as overheads and whiteboards toward multimedia presentations in which the presenter uses a PC combined with presentation graphics software such as PowerPoint (Microsoft), which is usually then projected to a larger display for the audience.

With the presentation source increasingly being digital, the natural assumption is that the recording method is digital as well. At present, however, the most common method for recording a presentation is via video camera, a well-established analog technology that, though reliable, has notable disadvantages that make it unsuitable for many institutions.

We will first introduce existing methods for presentation recording, noting their relative advantages and disadvantages. We will then present a method for recording multimedia presentations remotely and digitally that will result in excellent image quality. In addition, the method is virtually transparent to the presenter and the audience.


Existing Methods
Top
Abstract
Introduction
Existing Methods
Materials and Methods
Discussion
References
 
Client-Side Recording
Some presentation graphics software packages, such as recent versions of Microsoft PowerPoint, have limited functionality for recording the active presentation. In addition, some screen-capture software has the capability of recording all desktop activity. We group these as client-side recording because they share a common characteristic: the recording must be initiated and controlled by the presenter.

Client-side recording offers the advantages of low cost and excellent quality because of direct digital capture. Its disadvantages are that it requires intervention by the presenter to control the recording process. The presenter may not be technically savvy or may forget to initiate the recording. In addition, the recording software may lag or it may disrupt the presentation.

Video Camera
The video camera is currently the most commonly used method for recording presentations. Its use involves setting up a video camera and adjusting it to capture the presentation screen, presenter, or both.

The video camera offers the advantages of true what-you-see-is-what-you-get copy and the ability to capture pointing devices such as the beam from laser pointers. It is the most flexible of all methods, with the ability to record the presenter and the presentation as well as nondigital material such as overheads. Its disadvantages are its high cost. The results are highly dependent on equipment quality and technician skill because recording is sensitive to lighting, positioning, motion, and other disruptions.

Video Graphics Array (VGA) Split and Capture
This method takes the VGA output from any PC or laptop and sends it through a splitter device, feeding one output to the display projector as normal and another duplicate output to a separate PC that will record the video stream.


Figure 1
View larger version (20K):
[in this window]
[in a new window]
[as a PowerPoint slide]
 
Fig. 1 —Diagram shows hardware configuration of typical recording session. Note that recording PC has two inputs: network connection from presenter's computer feeds video; audio connection from sound system feeds audio.

 
The advantages of this method are that it offers excellent quality and is transparent to the presenter and presentation hardware. Presenters may use their own laptops without the need for any preconfiguration. The moderate to high cost because of the custom VGA-capture card required is a disadvantage. In addition, VGA cables are sensitive to noise, which limits the distance the recording PC may be placed from the presentation source.


Materials and Methods
Top
Abstract
Introduction
Existing Methods
Materials and Methods
Discussion
References
 
The new method described may be broadly divided into three separate steps: capturing, streaming, and recording.

Virtual network computing (VNC) is a mature, open protocol for the remote control of another computer. For our purposes, however, its primary feature of interest is its ability to relay screen activity from one computer to another over a network. In essence, the presenter (hereafter referred to as host) PC is preconfigured with VNC host software, which silently captures all screen activity and sends the visual data to another computer installed with VNC client software. The data may be streamed over any common computer network protocol, most notably Ethernet.

On the recording (hereafter referred to as client) side, a generic screen-capture program is used to record the screen data streamed from the presentation PC. For the audio, it is assumed an audio line-out may be easily tapped from the sound system of the presentation hall. The configuration layout is shown in Figure 1.

This method offers a number of advantages. The cost is low because all the software involved is open source and available for free, and there are no special hardware requirements. The recordings are of high quality because the visual data are captured directly off the host, avoiding any physical issues such as lighting. It is transparent because the presenter has no need to initiate or control any part of the process and does not need to be distracted by background activity. It allows for flexible positioning. Ethernet is relatively robust against noise, thus the recording PC can be positioned as far as 150 m away from the presenter. The recording PC may then be situated for convenience, for example, next to the sound mixing console or in a separate room.

The primary limitation of this method is the inability to capture the presenter's physical motions, particularly the common usage of laser pointers. The use of a computer pointing device such as a mouse will eliminate this problem, but a significant number of people find the mouse interface unwieldy for presentation use. In addition, specific software has to be preinstalled into both the host and client computers. In practice, this means that presenters are restricted from using their own systems unless they allow the technical staff to make the necessary changes.

Instructions for Use
Obtaining the software—There are three primary pieces of software that need to be obtained. VNC may be considered the backbone of the technique because it performs the remote screen capture and streaming. Many variations of the software exist for as many platforms; we used UltraVNC (Ultra@VNC Team), a free and open source implementation of VNC for Microsoft Windows (version 1.0, www.uvnc.com). Separate screen recording software is required on the recording computer. For this we used CamStudio, version 2.0 (RenderSoft, http://sourceforge.net/projects/camstudio), a previously commercial software package that has since been open sourced and made available for free. For recording the video, a particular issue of importance is the choice of codec used. Capturing in raw format is largely unfeasible because of the huge file sizes generated. We recommend CamStudio Lossless Codec, version 1.0 (RenderSoft, http://sourceforge.net/projects/camstudio/), an open source codec. The open source license in this case is particularly important because it avoids any future lock-in by proprietary vendors—a critical consideration if the recording is meant for long-term archives.

Installing the software—Install UltraVNC on the host PC. During the component selection stage of the installation, opt to install only the UltraVNC server and the UltraVNC video driver. In addition, ensure that UltraVNC is registered as a system service. Postinstallation, the only configuration needed is to set a password for the server.

Install UltraVNC on the client PC. This time, opt to install only the UltraVNC viewer. Next, install CamStudio with its standard settings. Finally, install the CamStudio Lossless Codec as follows: extract the files from the source archive to a temporary folder, right-click on the camcodec.inf file (its icon resembles a small gear on a notepad), and then click the install option (Fig. 2).


Figure 2
View larger version (35K):
[in this window]
[in a new window]
[as a PowerPoint slide]
 
Fig. 2 —Installation of CamStudio Lossless Codec.

 
Site preparation—For the physical site preparation, we assume a standard setup for presentations: the host PC connected to a multimedia projector and the presenter's microphone and host PC audio-out connected to the sound system.

To stream the video, the client and host PCs should be connected via a standard 100 baseT fast Ethernet cable, with both PCs configured to appear on the same network. For the audio, connect a standard stereo/mono line from the line-out of the sound system to the line-in of the client PC (Fig. 1).


Figure 3
View larger version (62K):
[in this window]
[in a new window]
[as a PowerPoint slide]
 
Fig. 3 —Configuration of CamStudio specifying recording window. In practice, we apply client-side scale factor of 0.75 to view window (1,024 x 768 -> 768 x 576) as trade-off between file size and quality.

 
Using the software—Once configured, usage on the host PC is totally transparent. All control is initiated and performed on the client PC. Instructions on usage follow a command 1 -> command 2 convention, in which each command between the arrows represents either a menu item or specific instruction to be executed in sequence. Start the UltraVNC viewer and connect to the host PC by specifying its IP address. Start CamStudio. The following commands are all executed within the program. Set up the recording window: region -> fixed region -> enable "fixed top-left corner" -> click the "select" button and drag the recording window to exactly encompass the UltraVNC viewing window (Fig. 3). Set up the video codec: options -> video options -> select CamStudio Lossless Codec version 1.0 under compressor settings. Disable local cursor recording: options -> cursor options -> hide cursor. Set up the audio source: options -> record audio from microphone. Set up the audio codec: options -> audio options -> audio options for microphone -> select PCM, 44100 Hz, 16 bit, mono. These settings are simply provided as a starting point. Different implementations may benefit from alternate settings, thus experimentation is strongly encouraged.

To begin recording, simply click the red record button on the far left of the CamStudio toolbar. The video will be saved into standard audio video interleaved (AVI) format when recording is stopped. The resulting video file size varies according to the settings used, but 100 MB per 15 min is typical. The raw video file may be archived as is or reencoded according to need. A sample session (http://radiology.um.edu.my/resources/misc/MR1.mov) is available on the Internet. It is a 30-min lecture on MRI safety issues reencoded in QuickTime (Apple Computer, www.apple.com/quicktime/) format. At 48 kbps (32 kbps video, 16 kbps audio), the video stream is feasible even for analog modem users yet retains excellent quality throughout.

During testing, both the presenter and recorder were equipped with identical 2.8-GHz Pentium 4 (Intel) systems with 256 MB of random access memory (RAM). UltraVNC is designed to minimize the impact on its host by allowing multiple customizations for trade-offs between capture quality and central processing unit (CPU) usage. Its impact on the host PC ranged from 10% using normal settings for a relatively static presentation to 30% using ultra settings for a full-screen, full-motion video embedded in the presentation). In all cases, the presenter was able to conduct the presentation smoothly with no apparent disruptions.

Limitations
It is difficult to quantify the temporal quality of the resulting video because both the host (UltraVNC) and client (CamStudio) may drop frames for performance reasons. In normal practice, we use a 10 frames per second (fps) capture interval, which appears to result in few or no dropped frames and is more than adequate for static slides and simple motion such as slide transitions and mouse pointer movement. Using full motion video as a worst-case scenario, we judged capture performance dropping as low as 3 fps. For presentations where such video clips are crucial to the message, it may be necessary to splice in the video file separately during postprocessing.

The interface on the recording side is currently quite unwieldy: pixel-accurate positioning of the recording window is necessary to capture the VNC stream. A side effect of this is the recording PC being forced into a dedicated role because any motion in the recording PC risks being recorded as part of the presentation video stream.

A design quirk in the CamStudio screen recording software results in a postrecord lag in which the software automatically encodes the audio stream after the recording session is halted. For a 40-min presentation, the postrecord lag may take up to 10 min depending on system performance. Any additional recording is impossible during the lag period, which could be a critical flaw in conference and similar situations in which many presentations are held end to end. In this case, alternative commercial software of similar functionality is available, which in our testing largely eliminated the lag (Camtasia Studio, version 2.1.2; TechSmith Corporation, www.techsmith.com).


Discussion
Top
Abstract
Introduction
Existing Methods
Materials and Methods
Discussion
References
 
VNC works over any standard networking protocol. There is no reason video streaming cannot be done wirelessly. The IEEE (Institute of Electrical and Electronics Engineers) 802.11b wireless standard, or Wi-Fi, has become ubiquitous in PCs and provides sufficient bandwidth for our purposes. Switching to wireless connectivity would allow even more flexibility in positioning.

In recent years, tablet PCs—PCs integrated with touch-sensitive screens—have been introduced to the mainstream market. These new systems present an interesting possibility for human interface. We predict that the use of such systems, combined with modified baton-like styluses, should help to overcome the inherent unease most presenters have toward traditional mouselike pointing devices.

Because VNC software is open source, it is open to modification and feature enhancements. Optimally, the VNC client should have the ability to record the video stream. This single feature would eliminate three limitations at once: no positioning is needed for the recording window; the recording PC need not have its viewing window open, thus it may be used for other needs; and the postrecord lag may be eliminated with proper recording implementation. At present, a small number of projects are exploring the idea of VNC stream recording, but none of them fulfills all the requirements.

In conclusion, a method for making high-quality digital records of presentations by using a mix of freely available software has been described. This method offers much potential for institutions on a budget as well as others seeking viable alternatives to current methods of presentation recording.


References
Top
Abstract
Introduction
Existing Methods
Materials and Methods
Discussion
References
 

  1. Tamm EP, Ernst R, Weems W. The virtual lecture: delivery of live and recorded presentations over the Internet. AJR1999; 172:9 -12[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?



This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Tan, L. K.
Right arrow Articles by Ng, K.-H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tan, L. K.
Right arrow Articles by Ng, K.-H.
Social Bookmarking
 Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS