Author: Abdullah Ruzni Zameek

Posted on

Week 2 : Photoshop Collage – Abdullah Zameek

As with most creative and artsy projects, my mind and imagination came to a bit of a standstill. During class time, I was messing around with a picture of the Shanghai skyline, so I thought I might as well try and incorporate some elements from that into my project. For the base, I used a picture of a place that I’m sort of missing these days – the NYU Abu Dhabi campus. I thought it might be fun to try and incorporate some elements from what I have seen in my brief time in China so far to a more familiar setting, creating a crossover of what seems to be two very different, but familiar settings. 

This was the final result: 

NYUAD walkway with elements from China

The source pictures that I used were the following :

Chinese Lanterns Bottle Opener ShanghaiPearl Tower Shanghai 

Chinese Street Food  NYU Abu Dhabi

The Process: 

The first step was to extract the necessary elements from the source pictures – the lanterns, the Pearl Tower, the Bottle Opener, a street food vendor and some clouds. This was quite easy to do with the the pen tool. However, if inspected very, very closely, it can be seen that the “Photoshopped” elements have very crude edges. This is because I did not create a smooth enough contour, but for the purposes of the image I wanted to create, it seemed sufficient. The next step was to add the necessary elements in. Once, I did that, I used the “History Brush Tool” to restore some of the elements from the original image that were being overlapped with the new image. This made the final image look more natural, and can seen at the intersections of the bridge and the middle of both the Pearl Tower and the Bottle Opener. The final step was to add a bit of haze and cloudiness to the scenary since that was the atmosphere I  was familiar with when walking around Shanghai. To do this, I added a solid color of gray to the image, and reduced its opacity down to a small percentage to give the image a hazy feel. Next, I reduced the saturation of the image, followed by the contrast. (I used this as a reference to do this last step).
And, this was the outcome!

NYUAD walkway with elements from China

Posted on

Week 02 Assignment: Case Study Research – Style2Paints- Abdullah Zameek

Style2Paints – An AI driven lineart colorization tool

One of the biggest “bottlenecks” in the comic/manga industry is the time taken for artists to find the perfect color schemes for their drawings, and to actually color them in. This makes creating a single chapter or volume of a particular work to be a long, tedious process. 
Developed as  a collaboration between four students at the The Chinese University of Hong Kong and Soochow University, Style2Paints is one of the first systems to colorize lineart in “real-life human workflow”. What this essentially means is that it tries to follow the exact same process that a human goes through when coloring in a picture. As the authors of the project described, the human workflow could be summed up as follows :

sketching -> colour filling/flattening – > gradients/adding details -> shading

The Style2Paints library mimics the same process, and generates 4 different, independent layers in the form of PSD files. The layers are :

  • Lineart Layers
  • Flat Color Layers 
  • Gradient Layers
  • Shading Layers

Style2Paints was inspired by past projects such as PaintsChainer[TaiZan 2016] and Comicolorisation[Furusawa et al]. These outputs of these two model, however, often contained artifacts and coloring mistakes. This is solved up some extent with the separate layer model that Style2Paints uses.
Having separate layers allows for an artist to adjust and fine-tune each layer before merging them to form the final picture. But, as described by the authors, the Style2Paints model is able to do most of that fine tuning for the user. The user inputs the lineart image, and three optional parameters which are hints(which color palette to focus on more, etc), color style reference images, and light location and color. 

The results generated by the model are classified as follows: 

Fully Automatic Result – When there is absolutely no human intervention.
Semi Automatic Result – When the result needs some color correction, the user can put in some color hints (clicks) to guide the model.
Almost Automatic Result – Semi automatic results with fewer than 10 human corrections.

The underlying technology behind this project lies in a two-stage convolutional neural network framework for colorizing. The first stage involves (called the drafting stage) involves an aggressive splash of color across the canvas to create a vivid color composition. This stage would contain color mistakes and blurry texture, would be fixed in the next stage where blurry textures are refined to smoothen the final output. This sort of model splits the complicated task of coloring into two, smaller tasks which allows for more effective learning models, and can even be used to refine the output from other models such as PaintsChainer.

A few images from the model have been attached below for reference. (It was reported that all the images below were achieved with fewer than 15 clicks)

I find this project to be interesting for multiple reasons. From a technical point of view, I like the fact that they approached the problem in the most “human” way as possible, i.e focus more on how a human would do it, rather than what would be most efficient for a computer to handle. Secondly, this model gives artists freedom to experiment with different colors at a faster pace. For example, they can try out different schemes using the model, and pick out on that works best in a short amount of time, as opposed to manually filling in the colors.  This would certainly help artists to create more exciting content in a shorter amount of time, which would ultimately be of benefit to the industry as  whole.

Sources :
The published paper can be found here
https://github.com/lllyasviel/style2paints

[P] Style2Paints V4 finally released: Help artists in standard human coloring workflow!
byu/paintstransfer inMachineLearning

Posted on

Response to The Medium Is The Message – Abdullah Zameek

The essence of McLuhan’s essay lies in the fact that the medium through which a message is conveyed has a direct impact on the impression that the message itself has on it’s audience. And, with any device that can be wielded by a human, comes a deep sense of responsibility in the sense that it can have both beneficial, or detrimental effects, depending on how it is used.  He says “Firearms are in themselves neither good nor bad, it is the way that it is used that determines its value.” This is such a timely statement in the present-day context because of how the Internet has evolved to become an information super-highway. And, it is important to take into account what sort of information travels across this highway, and to whom. For instance, various terror propaganda groups are notorious for using online social media to recruit members and spread malicious content. On the other side of the spectrum, there are the major corporations that favour certain content providers over others, creating an unequal environment where the ability to consume, produce and convey your content is determined by your financial and/or political prowess.  Technological advancements in certain sectors have the potential to create societal disparities as well. For example, certain education systems employ the use of modern multimedia and other technologies to deliver content, while others utilize traditional, analogue media such as blackboards and chalk as a medium.  And, in the field like education, the medium plays a pivotal role in determining how well the content is delivered and understood. Another sector that relies heavily on the medium of conveyance is the entertainment industry. With virtual and augmented realities becoming more mainstream and accessible, both producers and consumers alike, now have an additional choice when it comes to selecting how they wish to consume content. This also creates more opportunities for expression since newer media allow for more interaction, involvement and immersion than  ever before.

At the end of the day, like how the electric light in McLuhan’s essay was a medium for a spectrum of things, the Internet could also be considered to be very similar in that sense, and it is upto us to determine how useful or destructive it will be to society.

Posted on

Week 2 : Responses to “Long Live the Web” and “The Room Where the Internet Was Born” – Abdullah Zameek

The messages that Tim Berners-Lee conveys through “Long Live The Web” could not have been more timely. In any age where almost every service is available online, it is crucial to think about the way in which policies, standards and frameworks govern how we use the Internet. 
A lot of the modern problems regarding the Internet seem to stem from big corporations and other bureaucratic entities. Some of these problems are

  • Censorship – Not giving everyone with access to the internet the same information.
  • Net Neutrality – Favoring the services of certain entities over the others.

This sort of behaviour makes it seem like the Internet belongs to some organisation, rather than a free resource that belongs to everyone. However, unlike other free resources, everyone with access to it has complete privileges to determine how it is used and what it provides. It is collective effort that made the Internet what it is today – if not for the work of students, scientists and others in the early days of the Internet, we would not have the information super highway that we have today.  And, it is this sort of collaborative effort that needs to be revived and fostered as we move on into the future. The future of the Internet lies in open-souce collaboration, not closed-source, proprietary standards. Closed source standards create a hostile, competitive environment where companies are solely focused on outcompeting each other while not prioritizing user experiences and satisfaction.
While, this article had many resounding points, the following quote seemed to be quite questionable.
“Once you enter your data into one of these services, you cannot easily use them on another site.”
Privacy is one of the biggest concerns in the digital realm today, and the statement above seems to directly contradict the fundamental principles of digital privacy. If a user has provided their data to a service, it is expected that this service would not divulge this data to a third party, or use it for purposes other than what it is was collected for. If this sort of information-sharing was allowed, then there would be no control over who gains access to an individual’s data, and who would be accountable for the handling of such data. This is where the question of “Digital Human Rights” comes into play. It is high time that law-makers and technologists come together to create a framework upon which fundamental digital rights are agreed upon and put down in black-and-white. The ever-increasing presence of technology in the most intimate aspects of our lives warrants the need for such a framework. One such effort was the Online Magna Carta – an initiative by Tim Berners-Lee himself.
Once the Internet goes back to being the largest crowd-sourced, community-driven venture, would we be able to say that we have a free and open Internet for all.

“The Room Where The Internet Was Born” could be described as a historical account of the evolution of communication technologies, starting at the principles of “time-sharing” and going all the way to modern-day cloud-computing. The fundamental message behind this article could be that it is essential to understand the underlying inspirations that drive modern technologies and to understand and appreciate the roots of what we have today. As described, the ARPANET is simply an extension of the time-sharing concepts introduced by John McCarthy, and the Internet was built upon the ARPANET itself. After reading this article, I began to appreciate the current technologies that we have today much more, and it is essential that the humble origins of what we have today is recorded for the generations to come.