New Technology: Image-based Modeling and Rendering

IBRM which is known as Image-based modeling and rendering techniques have gained considerable attention in the graphics community because of their potential for creating very realistic images. The ability to capture subtle real-world effects and details related to the imperfections of the real world is one of the major benefits of these techniques which graphics researchers still do not know about how to model and render. Using images as both modeling and rendering primitives, image-based modeling and rendering entrances can help to relieve two necessary and long standing problems in computer graphics. First one is the need for simpler modeling techniques which is suitable for representing complex scenes, and the other one is the ever need for rendering acceleration.  By replacing conventional (geometric) models with image-based representations the former can be achieved. Rendering speedups are gained by detaching rendering time from scene complexity, and also by re-sampling pre-shaded images.

Basically, Image-based rendering (IBR) uses images as inverse to polygons, as modeling and rendering originals of rendering farm. While practicing, many IBR approaches assemble to image-geometry hybrids, with the assembling the amount of geometry ranging from each pixel depth to hundreds of polygons.  On the other hand Image-based modeling (IBM) means to the use of images to drive the rebuilding of three-dimensional geometric models. In spite of their potential, IBMR techniques are still in their infancy and several challenges which are still need to be overcome. With this tutorial surveys the state-of-the-art in image-based modeling and rendering techniques are discussing their strengths and limitations as well as enumerating some possible research opportunities. At the beginning, the techniques are classified as image-based rendering (IBR) or image-based modeling (IBM) and again purified according to their relative positions along the image-geometry spectrum.  For instance, finally, pure image-based approaches fall on the left part of the spectrum, whereas hybrid techniques are positioned according to the amount of geometric information required.

The goal of giving an intuitive description of the fundamental ideas behind most recent IBR and IBM techniques in a single document is very imaginative. Its hope is to give a useful reference for students and researchers who are interested in a solid introduction to the underlying principles of the young field. Unluckily, having space limitations, very few techniques had to be left out to guarantee proper coverage of the material treated here. The liking of the presented techniques took into account their fitness in a progression which helps the reader to gradually include the concepts and understand the field evolution. Techniques like as Image-based Visual Hulls, Light Field Mapping, Surface Light Fields, and Voxel Coloring expect proper recognition but could not be embodied in this survey for the lack of space.

Reasons behind the Trendsetting Shift of Interest in CG Industry since 2012

The multimedia industry has witnessed some trendsetting changes in the past few years. The most noticeable change that CG professionals saw in 2012 was the growing popularity of farm and cloud rendering. Eventually, a remarkable decrease in the demand for standalone local rendering processes for different multimedia projects was also witnessed.  Companies who were offering local renderingservices either changed their infrastructure to provide both local as well as cloud based rendering or lost in the competition and were closed down!

This distinctly visible trend of the multimedia industry signals at a few basics. CG experts tried to find out the reason behind this massive shift in interest from local rendering to farm rendering services offered by many companies.

Their research presented these concrete reasons that are at play behind this new interest towards cloud based rendering.

·        Ease of handling heavy workload

Multimedia companies literally struggled to provide their completed projects within the specified deadline in situations when they were burdened with heavy work load. One of the reasons which delayed the work was the slow rendering process available with local rendering services. So, the characteristic feature of cloud rendering where a cluster of computers are connected with each other on a single LAN has been a relief for them. Now rendering is much faster than what it was earlier. Hence, adhering to deadline is not a problem irrespective of the volume of work. 

·        No hidden cost

Maintaining an individual rendering machine has become a headache for most multimedia firms today. To be honest, the small sized companies simply cannot afford of spend a huge sum in maintaining a personal rendering machine. If we look at it from an apparent level, we might find that some local rendering machines come within the affordable limit of most CG companies. But there are hidden costs in it which lead to huge drainage of money every year!

The bulky rendering machine requires huge physical space for its installation.  The companies have to make separate arrangements for air conditioning and electrical requirements.  In addition to this, investing additional money for appointing separate workforce for taking care of its security and operating the machine also becomes a pain. The core business activities of the companies get hampered in the process.

But firms who have started outsourcing their rendering requirements to the cloud rendering service providers like Fox Renderfarm are gaining! There is no additional cost involved in the process and they are paying just for the service they require.

Top Four Trends in Cloud Computing during 2013

The year 2012 saw a massive transition of the multimedia industry from local rendering to cloud computing and rendering farm. The trend continued to be consistent throughout the year. But 2013 presented even more surprises for the multimedia professionals and CG artists! Requirements from cloud computing got magnified and the demands of the client companies got streamlined by the turn of this year.

Some of the top trends of 2013 in the sector of cloud computing has been listed below:

·        Shift of demand  in favor of hybrid cloud

The requirements of all the companies in the CG industry have got massively bifurcated. When cloud computing was initially introduced, companies were satisfied with a single cloud based rendering service provider. But the demands have become variegated now. So, companies prefer collaborating with a number of cloud computing firms at a time to satisfy their multiple interests.  Predictions are that this demand for hybrid cloud is here to stay for few more years. 

·        Demand for composite big data and cloud computing service

In order to become more resources, more and more companies showed preference for implementing big data analytics in their decision making processes.  So, the exclusive cloud providers could foresee big data as their chief rival in this competitive ambience. Majority of cloud computing services thus thought of upgrading their infrastructure in 2013. They aimed to win the loyalty of the target market by providing a composite service which will feature big data as well as cloud computing.

·        Stricter policies against cyber attack

Most small, big and medium sized companies showed their preference for cloud based rendering services. But a number of them were skeptic about the security of confidential data that are transferred via a common LAN. 2013 saw a really serious approach of cloud rendering service providers like Fox Renderfarm to provide highly secured data transfer.  A number of cloud providers incorporated the SMB technology in their infrastructure by the turn of the year. This highly technical security system will allow the client companies to scan the code of their source data and detect even the slightest attempts of cyber attack. Thus, client companies will no longer need to make any additional investment in arranging more expensive tools to detect any malicious presence in the source code.

·        Applications have become polished

Besides the fear of privacy of data, the client companies were also concerned about the performance of cloud computing. But 2013 was eventful for them as it exhibited the commitment of cloud providers to provide consistency in service. Thus, towards the end of this year the businesses feel assured that the system will never fail to perform in critical periods.

Rendering Technology-How It Started

Rendering technology has become indispensible for the multimedia industry today.  Graphics artists and multimedia specialists simply cannot think of creating 3D projects without rendering it repeatedly. But you will be surprised to know that the beginning of rendering can be traced back to the 60s!

Let us know more about the beginning of rendering and the first 3D object that was rendered.

The object that inspired

While discussing the beginning of render technology of render farm, mention needs to be made of Gordon Romney. He was one of the forerunners of 3D modeling and rendering who initiated the journey of creating computer generated objects and making it photo realistic.

The first object which inspired Romney to create a CG model was the ‘Soma Cube’.  It was a very simple rosewood box which was received by Romney in 1967. This rosewood box was developed by Piet Hein in 1933 and certified by the Denmark based brand ‘Skjode Skjern’.  The color, texture and visual effect of this Soma Cube were so outstanding that it encouraged Romney to develop its computerized form.

The basic framework

Romney discovered that the cube shaped rosewood box could be dismantled into seven parts: right screw tetra-cube, branch tetra-cube, left screw tetra-cube, ‘T’ tetra-cube and ‘L’ tri-cube.  He applied the same technique in making seven computer generated polygons that resembled dismantled parts of the original Soma Cube.

The steps that were followed

Romney created the seven polygons using a specific technique. The stepwise breakup of the technique is as follows:

·        Step one :

The light-pen driven Assembly Program was implemented for creating a cube-shaped meta-object from a list of pre-defined geometric objects. The hidden lines in the polygon were distinctly visible and it featured monochrome vector display.

·        Step two:

As it is known now, each of the components of the virtual Soma Cube had smaller units. At this stage, these unites were assembled together into standalone sub-objects.

·        Step three:

Now the color and texture of the components were decided and the necessary parameters were set according to that.  The pixel density, intensity of illumination, view point and other secondary parameters were also set.

·        Step four:

In order to place the CG object properly, the metrics needed for translation and rotation of the same were specified. The ‘Manipulation Program’ was used extensively to give a sequential action to the virtual Soma Cube.

·        Step five:

Now the Tektronix Oscilloscope was set by enabling film and camera options in it. The appropriate filter was also enabled. Separate renderings were done for the green, red and blue polygons in order to preview the rendered color image of each of the components of the Soma Cube. A final render was done when Romney was satisfied with the trial runs.

Although the process took long, now CG projects can be rendered faster using rendering techniques offered by firms like Foxrenderfarm.