We propose a technique that allows a person to design a new photograph
with substantially less effort. This paper presents a method that generates a composite image when a user types
in nouns, such as “boat” and “sand.” The artist can optionally design an intended image by specifying other
constraints. Our algorithm formulates the constraints as queries to search an automatically annotated image
database. The desired photograph, not a collage, is then synthesized using graph-cut optimization, optionally
allowing for further user interaction to edit or choose among alternative generated photos. An implementation of
our approach, shown in the associated video, demonstrates our contributions of (1) a method for creating specific
images with minimal human effort, and (2) a combined algorithm for automatically building an image library with
semantic annotations from any photo collection.
This submission includes the presentation and model files that were used in delivering a webinar on 12-15-05 that covered the topic of modeling Hybrid Electric Vehicles.
Hybrid electric vehicles (HEVs) have proven they can substantially improve fuel economy and reduce emissions. Because HEVs combine an electric drive with the internal combustion engine (ICE) in the powertrain, the vehicle?s kinetic energy can be captured during braking and transformed into electrical energy in the battery. The dual power source also means that the ICE can be reduced in size and can operate at its most efficient speeds.
In an electromagnetic cloak based on a transformation approach, reduced sets of
material properties are generally favored due to their easier implementation in reality,
although a seemingly inevitable drawback of undesired reflection exists in such cloaks.
Here we suggest using high-order transformations to create smooth moduli at the outer
boundary of the cloak, therefore completely eliminating the detrimental scattering
within the limit of geometric optics. We apply this scheme to a non-magnetic
cylindrical cloak and demonstrate that the scattered field is reduced substantially in a
cloak with optimal quadratic transformation as compared to its linear counterpart.
CRFsuite is a very fast implmentation of the Conditional Random Fields (CRF) algorithm. It handles tens of thousands sentences in merely one second.
In comparison to CRF++, CRFSuite yields substantially better efficiency performance
The version of CodeWarrior used in the ARM Development Suite
cannot export projects in makefile format. Manually producing
makefiles from CodeWarrior projects is not an easy task. The
aim of mcp2make is to make the conversion simpler and to require
substantially less effort.
mcp2make is supplied as a free-of-charge, unsupported utility.
obot control, a subject aimed at making robots behave as desired, has been
extensively developed for more than two decades. Among many books being
published on this subject, a common feature is to treat a robot as a single
system that is to be controlled by a variety of control algorithms depending on
different scenarios and control objectives. However, when a robot becomes more
complex and its degrees of freedom of motion increase substantially, the needed
control computation can easily go beyond the scope a modern computer can
handle within a pre-specified sampling period. A solution is to base the control
on subsystem dynamics.
The information technology (IT) revolution is surely coming in this century, just
as did the agricultural and industrial revolutions that have already so enriched our
lives. As the IT revolution progresses, it is expected that almost all social struc-
tures and economic activities will be changed substantially
When 3GPP started standardizing the IMS a few years ago, most analysts expected the
number of IMS deploymentsto grow dramatically as soon the initial IMS specifications were
ready (3GPP Release 5 was functionallyfrozenin the first half of 2002and completedshortly
after that). While those predictions have proven to be too aggressive owing to a number of
upheavals hitting the ICT (Information and Communications Technologies) sector, we are
now seeing more and more commercial IMS-based service offerings in the market. At the
time of writing (May 2008), there are over 30 commercial IMS networks running live traffic,
addingup to over10million IMS users aroundthe world; the IMS is beingdeployedglobally.
In addition, there are plenty of ongoing market activities; it is estimated that over 130 IMS
contracts have been awarded to all IMS manufacturers. The number of IMS users will grow
substantially as these awarded contracts are launched commercially. At the same time, the
number of IMS users in presently deployed networks is steadily increasing as new services
are introduced and operators running these networks migrate their non-IMS users to their
IMS networks.
Recent work has shown that convolutional networks can
be substantially deeper, more accurate, and efficient to train
if they contain shorter connections between layers close to
the input and those close to the output. In this paper, we
embrace this observation and introduce the Dense Convo-
lutional Network (DenseNet), which connects each layer
to every other layer in a feed-forward fashion.
