Ubicast EasyCast Studio User manual

EasyCast Studio user manual
version 2.6

GUIDE CONTENTS
Step 1. Setup (cables and connections)
Step 2. Recording a presentation
Step 3. Editing and publishing
Step 4. Viewing content
Step 5. Advanced settings

Step 1. Setup (cables and connections)

XLR/USB adaptor
XLR microphone
XLR
USB
Presentation PC
VGA splitter/scaler
VGA
EasyCast Studio
Network (RJ45)
Video projector
Power injector
Legend
Must be powered
Cable type
IP camera
Network (RJ45)
(optional)
VGA
VGA
EasyCast Studio setup
Setup time is approximately 10 minutes
for a mobile system
Setup is identical whether setting up a
mobile or fixed EasyCast Studio
The data input on the Studio is natively
DVI, a (supplied) VGA-DVI converter
will be necessary if using a VGA cable
from the presentation PC to the Studio
The use of a power injector enables you
to run only one network cable to the IP
camera. The cable going to the camera
should be plugged into the “Data &
Power - Out” plug on the power injector.

When using automatic speaker framing
(Autocam), the camera should be
placed :
•min. 3 meters away from the presenter
•ideally in the back of the room, in
order to cover the entire zone in which
the presenter will move.
Camera
Min. 3 meters
Camera placement for Autocam
Stage

Step 2. Recording a presentation

EasyCast boots directly into this home screen
– it is a dedicated Linux-based appliance. The
entire operating system is managed by
UbiCast and this is a big difference with
Windows-based lecture capture systems :
here the user can’t corrupt the system by
installing updates or any third-party software.
The entire user interface was created from the
ground up by UbiCast and is optimized for
touch screens.
Here, the user can :
•record a new presentation
•edit and publish existing media
•(1) press a single button for system wide
updating from drivers to the deep Linux
kernel (and consult the new version
changelog)
Press the (2) Record button to continue.
Home screen
1
2

When the user wishes to record a new media,
he must first chose a Profile. The default
profiles are device-based : you can either
record just (1) the computer and audio (data)
signal and audio, (2) the video camera and
audio, or (3) the computer along with video
and audio (to create typical Rich Media).
It is important to note that all of the system
options can be put into a Profile, allowing for
the creation of custom user-based profiles.
For example, here we could have just 2
profiles : HR and Communications. Selecting
HR tells the system that it needs to publish the
content to a specific location, in a specific
video format and burn a specific logo into the
video. The Communications Profile can be
entirely different and send the recording into
the client’s YouTube channel for example.
This enables the systems to be highly
automated.
If you wish to create profiles, please contact
your reseller or UbiCast directly – they can be
configured and added remotely.
Select (3) to continue.
Profile selection
1
2
3

The system has tried to find all the
devices and services requested by the
chosen Profile.
As indicated (1-3), we have found a
camera, microphones, a data signal
and (4) the Studio is connected to the
Internet, specifically to an EasyCast
MediaServer, allowing LIVE streaming to
a portal on www.ubicast.tv.
Security is built into the system here : if
no audio had been found, we could not
move to the next step, avoiding some
users to create content with no audio.
All devices have been found, press the
(5) Next button to continue.
Detected peripherals
1
2
3
4
5

Sources are previewed in zones 1 to 3,
respectively video, audio and data.
In order to start a live stream, the user must press
the Live button (4) then start a recording (5). This
is a security measure built into the system : if the
local network were to fail, at least we have a
recording of the content for later use. Also, it is
important to note that even if live streaming is in
low quality because the upload connexion is
weak, the recorded content is in full quality.
Before or during a manned recording, the user
can enter metadata concerning the recording by
tapping (6). This information can also be entered
later on of course.
During a recording, a user can also manually add
chapters (7) that will assist especially for media
trimming : for example during a conference,
press this button when a new speaker starts a
talk, this will enable to quickly cut out each talk.
For most users the only interaction here should
be just Start/Stop record.
Launch a recording, stop the recording and
press Export (8) to proceed. You can record as
many presentations in a row as the local hard
disk enables.
Recording screen
1
2
3
4
5
6
7
8

Step 3. Editing and publishing

This is the editing and publishing part
of EasyCast. You can access this part
of the system by tapping the “Publish”
button from the home screen.
Here we have the list of all media stored
on the system’s hard drive. Standard
systems come with a 2Tb hard disk and
can hold approximately 500 hours of
FullHD content.
Our previous recording is the media at
the top of the list. Select it and press the
play (1) button for playback. Below the
video, the captured slides will appear
as the presenter moved from one slide
to the next.
The first thing we may want to do is trim
the recording, to take out the first and
last few seconds. Tap the Trim (2)
button to move to the next step.
Media list - Publish application
1
2

Many EasyCast users wanted to import recorded
content into desktop editing software such as Adobe
Premier or Apple Final Cut only to cut out the first
seconds or the coffee break. To simplify the process
and because UbiCast’s technical developments are
client-driven, we simply developed the feature and
added it to the system.
The create an edit, simply move the time cursor (1)
to the desired starting point and press “Mark as
start” (2). Then move the cursor (1) to the desired
end point and press “Mark as end” (3). This creates
a segment that can be extracted by pressing the
“Extract” (4) button.
You can create multiple segments and extract each
of them as separate media files (data will stay
synchronized) or merge the multiple segments by
pressing the “Merge” (5) button before pressing
“Extract”.
Trimming is simplified by the fact that EasyCast
creates (colored) indicators on the timeline
corresponding to Microphone ON/OFF and Data
Connected/Disconnected events. So you can easily
cut the video when the presenter starts speaking, or
when he disconnects his PC before leaving for
example.
Note that this editing is non-destructive : the original
“parent” file will stay intact. Indeed, when you press
“Extract” you will see that the new media is created
as sub-media to the original video.
Media trimming (editing)
1
2
3
4
5

We previously trimmed/edited our
recording.
The second phase of editing here is
related to the Autocam (tracking)
feature in EasyCast. The system will
automatically reframe the 1080p video
we’ve created around the presenter for
dynamic content.
Tap the “Tracking” (1) button.
Media list - Publish application
1

There are multiples cases here :
1. You have never performed a
calibration of the system before the
recording, you’ve just set up the system
and pressed record. In this case, press
the “Configure and launch analysis” (1)
button and go to the next step.
2. You have already configured the
analysis zone (red rectangle). In this
case, just press “Launch analysis and
rendering” (2). The system will go back
to the Media list and start processing to
reframe (in this case skip the next 3
steps).
Note that pressing the “Calibrate
Autocam” (3) button will start a long
processing time that will find objects in
the room that might look like humans
and that should be removed.
Autocam - setup (1/3)
1
2
3

If you’ve pressed the “Configure and launch
analysis” button you will arrive on this screen.
By factory default, the analysis zone – the area in
which EasyCast will look for the presenter(s) – is the
entire video frame (zone 1). You can redefine this
zone by drawing a new rectangle. A few points :
•the smaller this zone is, the faster analysis time
will be
•to avoid detecting people sitting in the room, the
zone should contain only the presenter
•EasyCast looks for shoulder shapes (not faces),
so the suggested analysis zone should just
contain the presenter’s bust area (in this case for
example, zone number 2)
•the analysis zone should cover all the space the
presenter will be moving around in, so define wide
areas.
The default extracted frame’s size is defined by the
“Medium shot zoom factor (X)”. You can modify the
size of the frame using the “+” and “-” buttons (3).
You can also modify the vertical position of the
extracted frame (adapt to “unusual” body shapes)
by using the “Medium shot vertical offset” button (4).
Press “Apply” when you have the right settings. The
system will return to the Media list and start the
analysis process.
Autocam setup (2/3)
1
2
3
4

Once the analysis process is over,
press the “Tracking” button once more
in the Media list. You will arrive on this
screen, that shows you the detected
human(oid) figures.
In this example, a poster image has
been found (1). As you do not wish to
track this, select the rectangle and then
the “Ignore” button. The preview image
will then be of reduced opacity.
The same person has been found
twice, so select the two images (2 and
3) and press “Join”.
Press the “Apply” button to continue to
next screen.
If there is only one person found
(usually the case), simply press the
“Apply” button.
Autocam setup (3/3)
1
2
3

Zone (1) shows a preview of the frame that will be
extracted (green rectangle) from the original
video. By pressing the play button (2) you can
see that as the subject moves, so does the
extract.
On the video timeline (2), entry/exit of
“objects” (presenter walking in and out of frame)
is shown by vertical red bars. The green vertical
bars indicate a change in frame, typically going
from wide angle (original video) to tracked frame.
Segments (3), shown in grey, are created from
one green bar to the next. On each segment you
can :
•activate/disactivate the tracking of detected
objects (4) by simply tapping their image
•modify the zoom level, using the default
“Medium shot” or the “Custom shot” (frame
size defined in Settings)
You can manually create and delete segments if
you wish to manually change zoom levels during
media.
Press “Launch rendering” (6) to go back to the
Media list and start the actual extraction of the
green rectangle.
Autocam preview
1
2
2
3
4
5
6

When rendering is finished, there will be
a check symbol (1) to the right of the
video.
The video used for the media is
replaced by the tracked version (2). The
original video is always kept, so multiple
autocam versions can be made.
Our last editing step is chapters. Press
the “Chapters” (3) button to continue.
Autocam preview and Chapters
1
2
3

Chapters are the result of a human decision to
transform a slide into a chapter. To create a
chapter, simply give a slide a Title (1).
On the timeline (2), there are green and white
vertical bars.
•white bars indicate slide changes
•green bars indicate that a slide has been
transformed into a chapter
Users can either 1) not create chapters, 2) create
them manually by simply giving them Titles, or 3)
use an automation “trick” :
•most companies/universities have graphical
charters with titles always in the same slide
area (top center)
•if users place a (supplied) small chessboard
pattern (3) on any part of the slide, EasyCast
will find this pattern, “read” the title inside a
pre-defined zone (4) (can be modified by
drawing a new rectangle) using Optical
Character Recognition and automatically enter
it in the “Title” (1) zone. The slide is thus
automatically transformed into a chapter.
Here you can see the keywords that were found
on each slide (5) and view/edit the different types
of markers (6). These markers can greatly help
users when trimming content, for example those
that index audio on/off and data connected/
disconnected events.
Editing chapters
2
4
3
1
5
6