|
|
GStreamer Conference 2014 - speaker biographies and talk abstracts
Düsseldorf, Germany, 16-17 October 2014
Back to conference main page
Back to conference timetable
Over the Top GStreamer, Alex Ashley and Darren Garvey, YouView TV
The talk describes the work to enable over the top (OTT) TV services on the
YouView platform. YouView is a joint venture between BBC, ITV, Channel 4,
Channel 5, BT, TalkTalk and Arqiva as equal shareholders. The YouView
platform provides a seamless mix of TV channels and video on demand content
via digital terrestrial TV (DTT) and the Internet to over a million UK homes.
The software running on YouView set top boxes uses many open source projects,
including GStreamer.
The talk describes how we are extending our existing use of GStreamer as a
framework for supporting multicast delivery with support for HLS and MPEG DASH
for live streaming and on-demand media delivery. The talk briefly describes
how GStreamer is integrated in to the device and how it interacts with the
hardware audio and video decoders. The main focus is a description of the
additional features that we have added to hlsdemux and that we are adding to
dashdemux and qtdemux to support high performance trick play and encrypted
content. As part of this, there is a discussion on the challenges of providing
these changes up stream, given that we need to stay on a stable branch while
these elements are subject to rapid change on the git master branch.
Alex Ashley is the lead architect at YouView, where his primary technical focus
is in the area of IP media delivery. With almost 20 years of experience in
consumer electronics and digital television, in research and in development,
he has worked on projects from low level technologies such as MPEG encoding,
real-time file systems and WiFi, through to the launches of DVD-Video,
interactive TV, personal video recorders (PVRs) and Internet delivered
television (IPTV).
|
GStreamer's role in Tizen, Chengjun Wang, Samsung
This talk present a brief introduction to GStreamer's role in Tizen.
We will introduce how GStreamer is used in Tizen.
We would also like to discuss what where our main challenges, how
did we overcome them and present the audience with some of our
ongoing work (like our current closed caption work).
Tizen is the open-source operating system for all device areas.
"Tizen Profile" define requirements of SW and HW for each device category.
There are already various profiles such as Tizen Mobile, Wearable, TV.
Different profile have different scope and multimedia features.
With GStreamer, our goal is to build a high-integrated and scalable
multimedia solution for every Tizen profile.
Wang Chengjun is a Tizen GStreamer developer who works for
Samsung R&D China in Nanjing.
|
Bridging the divide between GStreamer and a largely C++11 codebase, Cort Tompkins, IPConfigure
GStreamer and C++11 can coexist wonderfully, but you need to use a light touch.
This talk will address how best to use C++ to create and manage GStreamer
objects, and how best to avoid the maintenance and interface nightmare of
creating your own "GStreamer++". Specific topics will include object lifetimes
for reference management, exception-safe error recovery, C++ objects inside
proprietary GStreamer plugins, and making your GStreamer plugins unit testable.
R. Cortland Tompkins is Vice President of Engineering at IPConfigure, a
developer of cross-platform, web-accessible video management and video analytic
solutions. Cortland holds a PhD in electrical engineering from the University
of Dayton and serves as an Adjunct Assistant Professor at Old Dominion
University in Norfolk, Virginia, USA.
|
GStreamer, Negotiate all the things, Edward Hervey (bilboed), BilboEd Consulting
This talk is in the same vein as last year's "time and synchronization"
talk, that is to make people understand some fundamentals of GStreamer,
its design and how it works (a bit) under the hood. It is *not* intended
to be uber-technical with plenty of code and API, it's more to make the
intent and design of GStreamer clear.
What makes GStreamer really awesome in this day and age ?
It's not only that you can simply describe, as an application, your
intention as a graph of processing to be done (along with convenience
element/bin to make that even easier), nor that you have all those
plugins available to do all this various processing, nor even that it's
available on all these architectures and systems... but most importantly
that GStreamer will do its best to end up with the most optimal path to
fullfill that intention given that intent, those various plugins, those
various platform/stream constraints.
So it comes down to negotiation, which 1.x has streamlined extensively:
- negotiate the elements that could potentially handle a new stream
(registry, pad templates + caps, klass...) and already with hints as to
which might be the most optimal (caps features)
- negotiate who will drive the processing and how (QUERY_SCHEDULING)
- negotiate a bit further, when there is a choice, what's the most
optimal capabilities (QUERY_CAPS, QUERY_ACCEPT_CAPS, ..) between various
elements
- negotiate the most optimal storage (QUERY_ALLOCATION and buffer pools)
to allow re-use of memory, avoid memory copies, ensure downstream
elements can use the fastest code-path, ...
- negotiate whether some processing is needed, whether some processing
can be delegated downstream, whether elements can provide more
information so that downstream elements can perform faster/simpler
(QUERY_ALLOCATION and GstMeta)
This will be explained along with:
- how elements indicate their preferred caps, features and pools (via
ordering: downstream offers, upstream decides)
- how elements reliably know if/when links should re-negotiated
(EVENT_RECONFIGURE), to always keep the most optimal processing paths
- how applications can influence/restrict the various negotiation
(forcing scheduling, restricting capabilites via capsfilter)
Edward Hervey has been contributing for over 11 years to GStreamer,
ending up there after starting the PiTiVi video editor and then
maintaining various components over the years. After having started
Collabora Multimedia in 2007 and ending up as the Multimedia Domain
Lead, Edward Hervey is now enjoying a sabbatical where he tries to mix
freelance consulting work while enjoying life.
|
GStreamer Continuous Integration, Edward Hervey (bilboed), BilboEd Consulting
As all big projects, GStreamer is now running a Continuous
Integration system. During this talk, we'll go over the initial
challenges, what is needed (and to be avoided), how the current CI
system and workflow works, and what the next steps and challenges are.
Edward Hervey has been contributing for over 11 years to GStreamer,
ending up there after starting the PiTiVi video editor and then
maintaining various components over the years. After having started
Collabora Multimedia in 2007 and ending up as the Multimedia Domain
Lead, Edward Hervey is now enjoying a sabbatical where he tries to mix
freelance consulting work while enjoying life.
|
GStreamer on Wayland Overview, George Kiagiadakis (gkiagia), Collabora
This talk will be about how GStreamer is getting along with Wayland in
1.4 and also about new features that are being worked and/or planned. It
will touch a bit of technical details, but it will also try to show the
advantages that using GStreamer and Wayland in combination has over
other multimedia display technologies. The talk will hopefully end with
a working demo.
George Kiagiadakis is a computer science graduate from the University of
Crete and a free software contributor since 2008. He got involved with
GStreamer in 2009 with a Summer of Code project in KDE, from which
QtGStreamer later emerged. Since 2010, he is working at Collabora where
he is assisting customers with the integration of GStreamer in their
products and researching new features.
|
Stereoscopic 3D Video in GStreamer, Jan Schmidt (thaytan), Centricular
GStreamer doesn't currently provide any explicit support for stereoscopic
content. This talk is about ongoing work to integrate strong 3D and multiview
support for GStreamer 1.x - to support technologies like 3D TV that are
already widely available, as well as an extensible framework for future systems.
At the GStreamer conference is 2010, Martin Bisson gave a talk about his
GSoC project to implement support for stereoscopic video. That work was for
GStreamer 0.10 and was never merged. It provided some inspiration and useful
code, reworked into a new design.
Jan Schmidt has been a GStreamer developer and maintainer since 2002. He
is responsible for GStreamer's DVD support, and primary author of the
Aurena home-area media player. He lives in Albury, Australia and keeps
sheep, chickens, ducks and more fruit trees than is sensible. In 2013 he
co-founded Centricular - a new company for Open Source multimedia and
graphics development.
|
Smart Properties for Pipelines, Jeongseok Kim, LG Electronics (cancelled)
Note: this talk has been cancelled and will not be presented this year.
This talk will cover simple property, query and event use cases for GStreamer
beginners. My colleague and I defined “smart-properties” to deliver value into
a pipeline.
The basic concept of smart-properties originates from a crash when setting
property in certain state. For example, someone defined “application-type”,
however, it must be set after hardware resource allocation. If not, the
property accesses invalid handle of hardware resource, as a result, it crashes.
This happened very frequently in my company. It’s a simple mistake, though.
Therefore, I will try to explain which one among them (property, query, and
event) is more appropriate for each situation. The talk will include revising
property, query and event, and a more simple approach to delivering parameters
into a pipeline with smart-properties.
Jeongseok Kim has worked for over 10 years as an embedded system software
developer. At LG Electronics he's a senior research engineer and also the
leader of GStreamer core team for WebOS.
|
HTML5, the Web and GStreamer, Luis de Bethencourt (luisbg), Samsung
This talk is about how GStreamer fits in the Web world. It will include an
explanation of the HTML5 Media API. How this works from the ground up. How
it is rendered by WebKit/gecko and how GStreamer fits in. The evolution,
problems faced and what lies ahead. Including some example web pages using
this API and code snippets to show and tell how it all works. The talk might
be a bit biased towards WebKit since it is what I have been at work lately,
but I've followed the notorious problems Firefox has had with GStreamer and
they need to be explained.
Luis de Bethencourt is a freedom-loving technocrat, who currently works for
Samsung's Open Source Group in London. He has always enjoyed programming
and playing around with video, so since he discovered GStreamer 5 years ago
he's been hooked. Originally from the Canary Islands, computers felt like a
door to the world. Luis saw open source software as the best way to enter
the innovative technology community, see how it all works, and become a part
of it. He enjoys being in front of the screen, behind the screen, Friday beers,
Sunday ice-creams, walks in the park, and people who read bios to the end.
|
Kurento: Creating A GStreamer-based Media Server For real-time Communication services, Luis López Fernández
In the area of real-time communications, state-of-the-art media servers
concentrate on offering just three types of features: Transcoding,
Multi Point Control Units (MCUs) for group communications, and Recording.
However, current trends on multimedia show that the market is demanding richer
capabilities. Following this, in this talk we introduce
Kurento an open source media server
written on top of GStreamer, which leverages GStreamer's flexible architecture
for exposing complex media processing capabilities including computer vision,
augmented reality, media blending, media filtering and more. During the talk,
we introduce the Kurento architecture explaining how GStreamer has been wrapped
and adapted to comply with the requirements of a full-featured media server.
We also introduce the agnosticbin: an element to ease dynamic pipelines
construction which makes simple the connection and disconnection of branches
in a pipeline while it is playing, avoiding deadlocks, non-negotiated or
not-linked errors. Thanks to this bin, Kurento has been able to expose a simple
API based on a “Lego-like” philosophy, where developers just need to
instantiate and connect their media elements without needing to worry about
low-level media details. We explain how this API has been adapted to browser
environments so that Kurento features can be directly consumed from WebRTC
enabled browsers using advanced JavaScript techniques such as Promises or
Generators.
To conclude, we discuss about the potential business applications of Kurento.
In a world where plain communications are becoming commodity, business models
based on plain recording and group communications are not really profitable.
For this reason, these advanced media processing mechanisms open new
opportunities given that they might provide differentiation and added value to
applications in many specific verticals including e-Health, e-Learning,
security, entertainment, games, advertising or CRMs just to cite a few.
Dr. Luis Lopez is associate professor at Universidad Rey Juan Carlos in Madrid,
where he carries out different teaching and research activities in areas
related to WWW infrastructures and services. His research interests are
concentrated on the creation of advanced multimedia communication technologies
and on the conception of Application Programming Interfaces on top of them.
The aim of such technologies is to simplify the development of professional
real-time communication services satisfying complex and heterogeneous
requirements. Dr. Lopez research ideas have generated more than 60 scientific
and technical publications and have been included into important research and
industrial projects including FI-WARE (http://fi-ware.org) and
NUBOMEDIA (http://www.nubomedia.eu). Currently, Dr. Lopez is leading the
Kurento.org initiative: an open source software infrastructure providing
server-side capabilities for WebRTC with features such as group communications,
computer vision, augmented reality, transcoding, mixing and much more.
|
OpenGL Desktop/ES for the GStreamer 1.4 pipeline, Matthew Waters (ystreet00)
OpenGL is a powerful API usually accompanied by dedicated
hardware. Equipped with GLSL, one can envisage complex (or simple)
filters, mixers, sources and sinks that transform, produce or consume
the typical video stream in extraordinary ways. This talk will provide
for an overview on the current integration state of GStreamer + OpenGL
(now with MOAR demos) and a look into the future of GStreamer with OpenGL.
Matthew Waters has only just started his hopefully long and rewarding
FOSS career after using Linux for the past couple of years. When he
isn't hacking on GStreamer's OpenGL support, he is attending University
and playing around with waveforms.
|
Implementing WebRTC capabilities for GStreamer: the case of the Kurento WebRtcEndpoint, Miguel París Díaz, Kurento
WebRTC is one of the most relevant technologies in the multimedia arena of the
last few years. Although WebRTC is still under standardization, it is currently
available on the browser of more than 2 billion users and hundreds of thousands
of videoconferences take place every day basing on this technology.
In this talk, we present our progresses for providing full WebRTC capabilities
to GStreamer, so that application developers can create GStreamer-based server
side infrastructures. We introduce the WebRtcEndpoint is a GStreamer media
element that implements the WebRTC protocol stack and encapsulates the
management of DTLS-SRTP, ICE, certificates, SDP options and many other of the
specific characteristics to interact with browsers (Chrome and Firefox at the
moment).
The WebRtcEndpoint is not a conventional element because it has been conceived
to be used through an API to manage easily the SDP negotiation needed to
establish the media exchange. It can be used as input and output in a GStreamer
pipeline given its capability to acts as source and as sink at the same time
receiving and sending media from/to a PeerConnection in the browser side,
another WebRtcEndpoint or another peer that implements WebRTC.
During the talk we will present the limitations we have found in GStreamer when
creating the WebRtcEndpoint and how we have get through them. We will also
introduce simple applications showing how the WebRtcEndpoint can be used in
coordination with other GStreamer media elements.
Miguel París is a software engineer and architect in the Kurento community.
|
State-of-the-art GStreamer on Android with Java and Scala, Nenad V. Nikolić
The Android platform is the mobile platform with presently the largest market
share which is not only used on smartphones and tablets but also the software
layer foundation in devices ranging from kiosks to laboratory instruments.
This talk describes the learning path, challenges encountered on it and a
working approach for developing Android applications for any device and
iteratively implementing the GStreamer integration. The focus is on integrating
video streaming capabilities using GStreamer while using Scala and Java
languages to implement the UI and application logic.
The main intention of this talk is to demonstrate how both GStreamer newbies
and experienced users can reliably and iteratively develop Android apps using
high-level languages like Scala and Java that interact with GStreamer native
code written in C. Another aim is to suggest how to gradually introduce
newcomers to GStreamer concepts that are relevant when integrating GStreamer
into Android application development.
Nenad is an independent software engineer living in Hamburg. Hailing from
Belgrade, Serbia, after 10 years of work experience with a variety of American,
West and North European clients both as employee and as a freelancer, he moved
to Germany in 2009 to work for XING on several web, API and search-related
products. Since 2013 Nenad is working as a freelancer. The majority of software
projects he was involved with typically have the Java platform at its core
where Java, Groovy and, most recently, Scala are the most commonly used
programming languages while other software components tend to be developed
in other languages as well like Ruby, Python, C/C++ etc. Inevitably, many APIs
have been not only used but also needed to be developed. A recent project has
brought Nenad to an interesting mixture of Android, Scala and GStreamer to
combine rapid test-driven development, continuous integration with performance
of (under tested) native code. Nenad holds an MSc degree in Computer Engineering
and Science and is fluent in 5 (natural) languages.
|
The Development of Video4Linux Decoder Support, Nicolas Dufresne (stormer), Collabora
This talk summarizes a year of development toward Video4Linux
Decoder support or more precisely modern Video4Linux API in
GStreamer Framework. This includes planar buffer support, tiled
video formats, Memory-2-Memory class of devices, addition of an
allocator, DMABuf, UserPTR, converter and more. And this is just
the beginning.
Nicolas is Senior Multimedia Engineer at Collabora, based in Montréal,
he was initially a generalist developer, with background in set-top-box
development. Nicolas started in 2011 contributing to GStreamer
Multimedia Framework, adding infrastructure and primitive to support
accelerated upload of buffers to GL textures. His work toward fully Open
Source general purpose use of accelerator in GStreamer continues today
at Collabora with the recent addition of Video4Linux accelerated
decoders and converters support, enabling playback of today's content on
Cotton Candy and the HardKernel Odroid U2.
|
Efficient Multimedia Support in QtWebKit on Raspberry Pi, Philippe Normand (philn) and Miguel Gómez, Igalia
Since the Raspberry Pi platform was introduced it dramatically
changed the picture of the embedded (affordable) micro-computer
market.
WebKit is a popular Web rendering engine developed by a wide
community of Open-Source developers including major actors such
as Apple, Samsung, Intel and Adobe.
The purpose of this talk is to explain how we managed to leverage
the hardware components of the Raspberry Pi to deliver acceptable
video rendering performance of a QtWebKit(2) browser using the
GStreamer 1.2.x APIs. We will explain how we integrated zero-copy
rendering in the multi-process WebKit2 architecture. Other
advanced use-cases such as video/canvas, video/webgl and
getUserMedia will also be presented.
Miguel Gómez is an experienced developer working for Igalia. He has been
working for 10 years on several linux based platforms, such as Maemo,
MeeGo and Tizen. He's familiar with Freedesktop, GNOME and Qt
technologies, and lately has been focusing on graphics rendering
optimizations on WebKit.
Philippe Normand is also working for Igalia. He has been working
for the past 5 years on WebKit and more specially integration of
GStreamer technologies within WebKit to support the latest HTML5
specifications.
|
GStreamer and Digital Television Support, Reynaldo H. Verdejo Pinochet (reynaldo), Samsung
Multi-standard Digital Television support is quite a challenge.
On one end we have a clear use case both independent users and
companies can exploit. In the other, we have a complex set of standards
and regional variations that had made it quite difficult for the
GStreamer development community to come up with a functionally
complete set of elements providing the base for ATSC/ISDB/DVB
application development.
At Samsung we have been working on filling some of this gap. This
work has been done with an upstream-first approach that has most
of our changes already available for the community to use and
build upon. Idea would be to present a brief introduction to the
challenge, what has been done, what's the current state of
the art and planed future work.
This talk aims to present a brief introduction to the DTV challenge, what
has been done in GStreamer, what's the current state of digital television
support with this multimedia framework and what does the future hold
for the community in terms of planned work.
Reynaldo Verdejo is a Sr. Software Developer working for Samsung's
Open Source Group. With a history contributing to FFmpeg and other
major multimedia projects, he has been working on multimedia-related
FOSS development for more than a decade, half the last one as a
GStreamer developer. After quite some time working with GStreamer
on Android he took a detour from platform enablement to work around
digital television (DTV) support, where he has been trying to get dvbsrc
in shape and extending its support to other broadcast standards.
Reynaldo lives in Penco, Chile with his wife, Catherine and their 3 daughters.
|
Cross-Platform WebRTC with GStreamer, Robert Swain (superdump), Ericsson
Ericsson Research has been actively developing a GStreamer-based real-time
communication framework for a number of years. At its core, it supports
configuration and setup of real-time communication sessions using RTP streaming.
Additionally the WebRTC API is supported and has been demonstrated in the
Bowser web browser for iOS that was previously available on the App Store.
Browser plugins and native applications across Linux, Mac OS X, Android and
iOS are also supported.
The framework, called OpenWebRTC, was
released as Open Source at the beginning of October this year along with the
source code to Bowser for iOS.
Robert Swain, Senior Researcher at Ericsson Research. Robert has been working
on RTC for less than a year, on GStreamer since 2009 and on open source
multimedia software in various capacities since around 2000.
|
Trick Modes in GStreamer, Sebastian Dröge (slomo), Centricular
Faster and slower than real-time playback and reverse playback, i.e.
"trick modes", has been gaining importance in the industry lately. While
GStreamer has had the necessary support for many years already, it is
a rather complicated topic with many pitfalls and incomplete implementations.
This talk will explain the theory behind trick modes in GStreamer, the
different approaches for implementing them and provide some case studies
for the common cases of local files, adaptive streaming (HLS/DASH) and
server side trick modes (RTSP/DLNA).
Sebastian Dröge is a free software developer and one of the GStreamer
maintainers and core developers, and also contributing to many other
free software projects. While finishing his computer sciences degree, he
started working as a contractor for Collabora and stayed there until
2013 to work on GStreamer and related technologies.
Nowadays Sebastian is working at Centricular, a company providing
consultancy services around GStreamer and Free Software in general.
|
A New Tracing Subsystem for GStreamer, Stefan Sauer (ensonic), Google
GStreamer applications process large amounts of data in multiple threads.
As the processing is time-critical running a classic debugger to investigate
issues does not work well. The current practice is to use extensive textual
logging. The new tracing subsystem introduces hooks, plugin and structured
data logging. With this we can write tools to understand the logs,
automatically analyse them and give a better presentation of the data.
Stefan is a software engineer working for Google on build infrastructure tools.
In the past he was working for Nokia on the multimedia stack used on their
maemo/meego internet tablets and phones. In his free time his is contributing
to GStreamer, other GNOME projects (e.g. gtk-doc) and working on his music
editor buzztrax. He has a PhD in Human Computer Interaction from the Technical
University of Dresden/Germany. Stefan now lives in Munich/Germany with his
wife and his two kids.
|
HLS, DASH, MMS: Adaptive Streaming Formats in GStreamer, Thiago Sousa Santos (thiagoss), Collabora
Adaptive formats allows content distributors to provide multiple bitrate
options for media streaming for the same content while reusing the already
available file transfer infrastructure on the web. Clients consuming adaptive
streams can select the most appropriate bitrate for their context and switch
during playback if needed when, for example, the network speed fluctuates.
Implementing GStreamer plugins to handle adaptive content raised new challenges
and should pose as an interesting study case for the community.
GStreamer currently supports all three major adaptive streaming formats
(HLS, DASH and Smooth Streaming) for consumption and playback, it already has
support for generating HLS streams and support for DASH is on its way.
This talk will go over the updated status of these features and detail the
solutions involved.
Thiago Santos started working with GStreamer in 2007 while still taking his
Computer Science degree course in Brazil. Joining Collabora Multimedia
in 2009, he has become a GStreamer developer ever since and is interested
specially in complex elements like playbin, camerabin and, recently,
adaptive formats support.
|
Validate your elements' behaviour, Investigate your bugs and avoid regressions with gst-validate!, Thibault Saunier (thiblahute), Collabora
GStreamer Validate is a debugging tool that traces the behaviour of each
element in the pipeline and report any behaviour that is not correct. On
top of that it implements the notion of scenarios, allowing to execute
actions on the pipelines (seeks/trick modes, set properties during
playback, set pipeline states... basically anything can be done). And
there is also a simple testing framework allowing to describe tests,
execute, and create reports for these tests.
This talk will focus on how useful this tool can be for developers to
simply reproduce and thus fix bugs through live examples. It will also
explain how it is currently being deployed as an integration test tool on
the GStreamer Continuous Integration infrastructure.
Thibault Saunier started working on GStreamer through the Pitivi project in
2010. He has since then been working in many areas of the framework. Thibault
maintains GStreamer Editing Services, GNonLin and gst-python and more recently
he started to dedicate time building an integration testing infrastructure
for GStreamer: GstValidate.
|
GStreamer State of the Union, Tim-Philipp Müller (__tim), Centricular
This talk will take a bird's eye look at what's been happening in and around
GStreamer in the last twelve months and look forward at what's next in the
pipeline.
Tim Müller is a GStreamer developer, maintainer, and backseat release manager.
In the past he worked for Fluendo and co-founded Collabora Multimedia.
Last year he joined forces with GStreamer legends Jan Schmidt and
Sebastian Dröge and started Centricular Ltd, a new Open Source consultancy
with a focus on GStreamer, cross-platform multimedia and graphics, and
embedded systems. Tim lives in Bristol, UK.
|
MPEG-H 3D Audio: the next generation audio compression standard, Tomasz Żernicki, Zylia
This talk will cover the design and specification about MPEG-H 3D Audio
compression standard.
MPEG-H 3D Audio is a combination of two standards: multichannel MPEG-D
Surround and audio object coding MPEG-D SAOC (Spatial Audio Object
Coding). MPEG-H 3D Audio standard should be finalized at the end of 2014.
The main goal of such standard is to provide a coding compression scheme
suitable for sound reproduction where the number destination loudspeakers
varies from one to even 22.2. At the same time MPEG-H 3D Audio will
provide to user the possibility of interactive sound source manipulation
in the sound field. Total bitrate varies from 256 up to 1200 kbps for 22.2
channel material.
The first part of the talk will provide a high level overview of MPEG-H 3D
Audio, with brief descriptions of included coding and rendering tools. The
second part of the talk will provide insight on possible use cases and
market adoption of this technology.
Tomasz Żernicki is a co-founder and a Senior Research Scientist at
NextDayLab and Zylia R&D. companies are focus on applied research in the
field on multimedia and networking. His professional interests concentrate
on: audio and video compression algorithms, spatial sound processing,
audio bandwidth extension, parametric sound representation and modeling.
He received PhD degree from Poznan University of Technology (Poland) in
the field of Electronics and Telecommunications. As an audio expert he
takes an active role in the Moving Picture Expert Group (MPEG)
standardization committee work.
|
Image Conversion, Wim Taymans (wtay), RedHat
How images are represented in computer memory and how to convert
between one representation to another is an often poorly understood
topic. This talk will cover how video is captured, subsampled, color
converted and stored. We will talk about format and colorspace
conversions and how to upsample various subsampling formats. We will
also see how interlaced formats are stored and converted. Finally
we go over all the steps needed to convert video from one format
to another an we'll also cover what shortcuts are usually taken to
make this more practical in multimedia applications.
Wim Taymans has a computer science degree from the Katholieke
Universiteit Leuven, Belgium and decades of software development
experience. He co-founded the GStreamer multimedia framework in
1999 and is the person behind much of the current design.
Wim Taymans is a Principal Software Engineer at Red Hat,
responsible for various multimedia packages and pulseaudio.
|
How to Control Playbin Playing in a Limited Resources Environment, Wonchul Lee, LG Electronics
Multiple pipelines can run on an embedded system. It is becoming
mandatory feature for support some apps and scenarios.
But what if a hardware decoder is fully occupied by a certain pipeline,
and has to play additional media? The pipeline needs to be managed for
preventing lack of resources and, for example, prohibit construction of
a new pipeline until the old one is released.
In this talk I share my experience about how to control a playbin pipeline
and manage resources by using pad probes.
Wonchul Lee has been working for LG Electronics since January 2011 as a
researche engineer. He joined the media (GStreamer) project in the middle
of 2011 for LG TV. He mainly develops GStreamer for LG specific features
and maintain stability of pipeline.
|
GstHarness again - A follow up, Håvard Graff, Pexip
Where is GstHarness now? Tips and Tricks on the use of this new testing tool
for GStreamer.
Håvard Graff has worked with GStreamer professionally for 7 years in
Tandberg, Cisco and now Pexip. Developing video conferencing systems
like Movi, Jabber Video and Pexip Infinity using GStreamer as the
backbone. Was instrumental in premiering GStreamer in the AppStore.
Still pretends to be a musician with programming as a hobby.
|
Pexcision - Real-time media-detection testing-framework, Håvard Graff, Pexip
Automated testing of large complex videoconferencing scenarios is tricky. How
to test videomixing and audiomixing? How do we ensure a mixed stream contains
a particular input stream or set of streams at any given point in time. We
talk about how we are using special GStreamer sources and sinks to achieve
this.
Håvard Graff has worked with GStreamer professionally for 7 years in
Tandberg, Cisco and now Pexip. Developing video conferencing systems
like Movi, Jabber Video and Pexip Infinity using GStreamer as the
backbone. Was instrumental in premiering GStreamer in the AppStore.
Still pretends to be a musician with programming as a hobby.
|
Videocodec development with GStreamer, Stian Selnes, Pexip
Opportunities and challenges with developing videocodecs in a GStreamer
context, especially for low-latency coding in lossy environments. Examples
on how to track down rare bugs, tools and tricks for testing and development.
Stian Selnes is a software engineer developing video conferencing
systems at Pexip. He has been working on GStreamer, video codecs and
other types of signal processing for 7 years at Pexip, Cisco and
Tandberg.
|
Porting Pexip Infinity to 1.x - Live pipeline challenges, Stian Selnes, Pexip
Discussing our port from 0.10 to 1.x with the benefits and problems that
came with it.
Stian Selnes is a software engineer developing video conferencing
systems at Pexip. He has been working on GStreamer, video codecs and
other types of signal processing for 7 years at Pexip, Cisco and
Tandberg.
|
Lightning Talks
Lightning talks are short talks by different speakers about a range
of different issues. We have the following talks scheduled so far
(in no particular order):
- Time synchronization in multipoint audio streaming using GStreamer in 3DAudioSense system · Tomasz Żernicki, Zylia
- Testing Video4Linux Applications and Drivers · Hans Verkuil, Cisco
(the long version of this talk can be seen on Tuesday October 14 at 11:15am as part of ELCE)
- Automatic audio-video synchronisation measurement using GStreamer and OpenCV · Florent Thiery, UbiCast
- Using GStreamer to set up Raspberry Pi as a surveillance camera · Vivia Nikolaidou
- Low Power Audio pipeline with PulseAudio · Manohar Babu K, LG Electronics
- Dynamic switching of music playback on hardware/software pipelines for Low power Audio · Biju Malayil, LG Electronics
- The Linux DVB libdvbv5 library for digital TV applications · Mauro Carvalho Chehab, Samsung
- gnonlin? it's dead, jim (a new design for dynamic pipelines) · Mathieu Duponchelle
- Reducing the loading time with playbin · Hyejin Choi, LG CNS
- Secure RTP · Wim Taymans, RedHat
- Zero-copy inter-process video with FD passing · William Manley, stb-tester.com
- Special Interest Groups? · Edward Hervey
- RTMP - Challenges with testing and interop · Håvard Graff, Pexip
- Your talk here?
Lightning talk speakers, please export your slides into a PDF file and
either send it to Tim by e-mail (you should have gotten an e-mail by me about
your lightning talk) or have it ready on a usb stick before the start of the
lightning talks on Thursday. The idea is that everyone uses the same laptop
so that we don't waste time hooking up laptops to the projector and configuring
them. There is no particular order or schedule for the talks. When a speaker
is called up, we will also mention who is up next. Every speaker has up to
7 minutes for their talk. There will be a countdown timer running, and there
will be some music playing towards the end so the speaker knows they have to
wrap up. If you don't want to use up the full 7 minutes, that's fine as well.
We will try to squeeze as many talks into the Thursday slot as possible, and
then finish with the rest on Friday.
|
|
|
|