Post-processing 3D renders with visualisation artist Piotr Kasprzak

Piotr Kasprzak of Infrared CGI is a visualisation artist who specialises in creating 3D graphics for architectural and interior visualisations, product advertising and 3D animations.

Here we learn more about his workflow, the challenges involved in achieving photorealistic renders and how he uses Affinity Photo for post-processing his images.

Can you explain a little bit about what you do and how your company Infrared CGI came into existence?

Infrared CGI is a multi-purpose visualisation brand, mainly focused on architectural and product visualisations and CG animations. It came into existence about two years ago.

What range of expertise do you bring to your clients?

Our range of expertise is really broad and it depends on the clients’ needs. For architecture clients, we create still and motion images, animated GIF’s, 360 virtual tours and full CGI animations. For product advertisement, clients prefer still images, animated GIF’s and short animations.

Can you tell us about the software you use for your work?

The workflow mainly relies on 3ds Max and Corona Renderer, and Affinity Photo for post-process.

Can you give us an overview of how you create a 3D visualisation from start to finish?

Sure. After receiving files (blueprints/specs) from the client, I make a couple of “clay renders” (3D renders without any textures applied) in order to show to the client possible camera views for the particular image. Clients usually make small changes to the proposed views. After applying changes and the clients acceptance, I start to work on the draft image, in order to present the mood of the image, colours, composition. Usually, the mood of the image is determined by the client, but sometimes I make a second version of the image, in order to show to the client a new and fresh look, different from the original one. After the series of changes the client receives final, full post-process, high-quality images.

Rivers Walk project: final post-processing steps in Affinity Photo
Typically how long does it take to create an image?

Usually three to five working days for the first draft, and then one to two working days for applying changes requested by the client and delivery of the final image.

Why do you choose to use Affinity Photo for post-processing your images?

Speed, full 32-bit depth support, non-destructive workflow and Photoshop plug-in compatibility! A huge advantage is also the price and perpetual licensing.

How well does Affinity Photo work alongside your other software?

It works great! Thanks to full 32-bit support, I can render the EXR files in my 3D software, import them into Affinity Photo and have full control over what I am going to do.

What are your typical post-processing steps for creating photorealistic renderings?

As I mentioned before, I am using 32-bit in post-processing.

I put all the render layers into Affinity Photo and compose them into beauty render. This is the usual starting point. The creative process starts beyond that point.

I start to search the image’s mood by replacing the sky, then I make adjustments to render elements in order to better fit the overall feeling of the image. Then I start to colour correct the image.

At the end of the process, I usually add people and vegetation cutouts, colour correct them in order to blend better with the image and the last step is to add slight contrast and sharpen the final image a bit.

The final image

Check out this timelapse video which shows the post-processing stage in full:

What would you say is the most challenging aspect of 3D visualisation?

There are a lot of different challenges during production. I think the most demanding is to achieve true photorealism in visualisations.

Does your process differ between architectural, interior and product renders?

Yes. Interior and product visualisations need to be photorealistic. Architectural visualisations can be more stylised, which means they can be more stylistically and artistically defined. To achieve photorealism there are certain rules that must be followed. If something is missing, you can tell that something is wrong and the image doesn’t look quite right.

Where do you look for inspiration for your work?

A big part of inspiration comes from the world around me. I am also inspired by Frederic Edwin Church paintings, who was a great landscape painter. Other 3D artists’ works are also an inspiration to me. Behance and CG Architect are good inspiration sources.

What would be a dream project for you to work on?

Visualisations for SpaceX or NASA. Visualising Martian architecture would be a dream job for me!

What would you suggest to someone wanting to become part of this industry? What are the essential skills to focus on when first starting out?

Skills related to the 3D industry are essential, ie. 3D modelling and texturing. Basic knowledge of compositing and colour theory are important too. The ability to work in a team, time management, being responsible for your work, those are all important personal skills.

How soon will visualisations be indistinguishable from real photos? Or do you think that we have already reached that point?

For persons unrelated to the 3D industry, visualisations can be indistinguishable from real photos.

Personally, I think there is a long road for reaching a flawless photorealistic look. Render engines are very good at mimicking the real world, but there are a lot of things that need to be done if we want to achieve a successful photorealistic look. If just one of these things is not perfect, the whole image looks fake.


You can see more work from Piotr of Infrared CGI at infraredcgi.com, Behance and Instagram.

To find out more about EXR Support in Affinity Photo check out our information at affinity.help.