Last updated on December 8th, 2023
Topaz DeNoise AI is a dedicated noise-reduction software developed by Topaz Labs. It uses artificial intelligence to reduce photographic noise rendered by a camera. High-noise photos can now be quickly recovered — instead of discarded — in post-processing. I’ll explain how the software works in this Topaz DeNoise AI review.
Related: You may also be interested in the Topaz Photo AI review with noise reduction.
There’s nothing worse than taking a photo and only realizing the excess noise it contains when viewing it at home. Oftentimes, these noisy images are decided to be useless to be shown online or printed and framed. This can be worse if you’re a professional photographer — your business and brand depend on producing high-quality images. The good thing is photography software technology has come a long way, including noise reduction abilities.
Hence, we got you covered with noise reduction in this Topaz Denoise AI review.
Summary of My Topaz DeNoise AI Review Version 2
Before I go into more detail in this Topaz DeNoise AI review, let me say that Topaz DeNoise AI works. And it more than works — it’s exceptional considering the lack of user input or required adjustments.
What is Topaz DeNoise AI?
Topaz Labs developed DeNoise AI to do only one thing: reduce noise in an image without losing detail. The software reduces noise efficiently without the user needing to play around with the settings. This saves photographers quite a bit of time.
Before we go any further in this Topaz Denoise Review, we first need to understand the basics of image noise.
What is Noise?
Have you ever zoomed in on an image and noticed the grain-like dots? That’s noise unless you’re zooming in on sand.
There’s an entire scientific explanation for what causes noise which I won’t cover here. But I’ll mention that digital photographic noise is caused by the image sensor’s response to light. Heat, electricity, and the image sensor’s illumination all play a role in creating unwanted noise.
And since the aperture and shutter speed controls the amount of light reaching an image sensor, they also influence the amount of noise in an image. More importantly, ISO, which is the image sensor’s sensitivity to light, usually plays a bigger role in digital noise.
Cameras with small or cropped image sensors tend to produce noisier images because of the size of the pixels in the image sensor. This is one of the reasons many photographers prefer shooting with a full-frame camera.
During the post-processing workflow, noise is broken down into two categories:
Luminance Noise
Luminance noise is the most common type of noise you’ll notice first in images. Its obvious almost-white color is spread across an image, making it look like you don’t know how to photograph. It’s more noticeable in the highlights of an image, such as the sky or in dark shadows.
Chroma Noise
Chroma noise, also known as color noise, is uneven color transitions. It looks like small patches of color that were smeared. Red seems to be the most common color in chroma noise and is highly noticeable in dark images. But green and purple color noise shows up now and then.
Usage Cases for Topaz DeNoise AI
Several scenarios can arise when noise reduction software is required. This applies to professional photographers, enthusiasts, and even non-photographers taking a family portrait.
- long-exposure photography
- painting with light
- photographing light streaks
- astrophotography
- media photographing current events without flash
- wedding photographers not using flash
- wildlife photographers where using flash is often prohibited
- street photographers that don’t use flash at night
- mobile phone photographers
- owners of APS-C sensor cameras
- owners of micro four-thirds sensor cameras
- owners of entry-level compact cameras
Post-Processing Workflow
For optimal results during the post-processing workflow, reducing noise at the beginning is recommended. This is because you don’t want to add contrast, saturate, or sharpen the noise in your image, making it more difficult to remove later.
How Topaz DeNoise AI Works Summary
High-end cameras, optimal camera settings, and ideal scenarios produce minimal noise. Cases like these don’t require noise reduction in post.
Unfortunately, ideal scenarios, like public transportation during rush hour, don’t come by often.
This is where noise-reduction software or tools like Topaz DeNoise comes in.
As mentioned earlier, Topaz DeNoise uses artificial intelligence to reduce noise in an image. According to Topaz Labs, it’s the world’s first AI-powered noise reduction software. That’s a bold claim.
I know the term AI is thrown out there a lot, and it’s used as a marketing term. Maybe it shouldn’t be called artificial intelligence since that’s what we expect software to do. But the term is here.
Getting back to how the software works, DeNoise AI uses machine learning to determine what’s noise in an image and what’s not.
They did this in a few steps.
They used an original photo with minimal to no noise.
Then they digitally reproduced the noise in the original photo. You may be asking, why not take a photo with a lot of noise in it already? The reason is they needed the original photo for comparison or as a reference photo.
Next, they inputted the noise-altered photo into DeNoise and checked its output file against the original photo. Either it’s a success or not.
They repeated these steps millions of times, made updates where needed, and that’s how DeNoise’s machine-learning technology works.
Topaz DeNoise AI Features
One of the good features of Topaz Labs’ software is that when you open them, a product tour pops up. It’s simply a guide to give you a basic understanding of how the software works. Most of the time, it’s good enough to go through the product tour without referencing an external online manual.
I’ll break down some of the important tools in DeNoise AI.
First, when you open an image in DeNoise AI, a default noise reduction setting will be applied. It’s called the DeNoise AI model and will usually have a default value of 15 for both the Remove Noise and Sharpen sliders.
There are two different noise reduction models:
DeNoise AI – Preserves more detail but works with fewer images.
AI Clear – Works with more images but may preserve less fine detail.
Both the DeNoise AI and AI Clear models have a Remove Noise and Sharpen setting. However, DeNoise AI uses sliders where you can make incremental adjustments. On the other hand, AI Clear uses presets of low, medium, and high.
Nonetheless, if you need to tweak your noise reduction and sharpness settings, both models will allow you to do it.
Photographers can also use the Auto-detect settings for the software to detect the amount of noise in the image and automatically apply the most optimal settings.
To further refine your image, you can also recover details and reduce color noise in both models.
Sometimes it’s challenging to see noise in low-light images. Thankfully, Topaz Labs also added a Brighten feature. As the name suggests, it increases the brightness of the image and allows you to see noise easier.
DeNoise AI allows you to compare your before and after images in three ways: a single view where you can click and hold the mouse to see the original image, a side-by-side view, and a split view that includes a vertical slider.
If you prefer doing selective noise reduction, you can apply a mask to DeNoise AI. The mask works in the same way as Lightroom’s mask with edge-aware.
As much time as Topaz DeNoise saves you with its AI technology, it can become cumbersome to reduce noise on several images. To make the noise reduction workflow faster, the application includes batch processing. All you’ll need to do is drag multiple images into DeNoise AI or open them through Windows Explorer or Mac Finder. Batch processing will allow you to leave the noise reduction model at its default of DeNoise AI or adjust it manually.
Once you’re happy with the noise reduction settings, you’ll have to click on Save Image to render the image. You’ll have the option to save your image file as DNG, JPEG, PNG, or TIFF.
Though I used some RAW files in my Topaz DeNoise AI review, it’s better to use TIFF instead because the software uses third-party libraries to render RAW files. Conflicts can arise.
All major RGB color profiles are supported in DeNoise AI: ProPhoto RGB, Adobe RGB, sRGB, Apple RGB, and Wide-Gamut RGB. CMYK is not supported at this time.
Plugins
DeNoise AI comes packaged with plugins for Photoshop, Lightroom, and Topaz Studio 2. There is no extra work to install the plugins. When you install DeNoise AI, the plugins will install automatically. You might have to restart Photoshop, Lightroom, or Topaz Studio 2 if they were running to have the plugins take effect.
How to Access DeNoise AI Plugins
- From Photoshop, DeNoise AI can be accessed from the Photoshop Filter menu.
- In Lightroom, DeNoise AI can be accessed by right-clicking on the thumbnail photo and then go to edit in.
- From Topaz Studio 2, you can access DeNoise AI by going to menu Filters → Plugins.
Summary of the Test Cases for DeNoise AI Version 2
Initially, in this review, I tested version 1.3.3.
The latest version of DeNoise is 2.2.2, which I’m using now to update this post.
The first image I tested was that of a pelican in low light. The image was photographed with a Canon 6D full-frame camera at 1/160, f/2.8, ISO 1000.
One of the noise issues is that you don’t always notice it until you zoom in. As is the case with my pelican image. Most photographers would know that an ISO of 1000 and a slower shutter speed than normal of 1/160 will produce some noise.
How to Use Topaz DeNoise AI
I took my DNG image file and dragged it into Topaz DeNoise AI. You may be wondering why it’s a DNG file. Well, during the import process of my CR2 Canon RAW files into Lightroom, I converted them to DNG to reduce disk space. In this case, I decided against using a TIFF to see how the image renders.
I used the Brighten tool to see the noise more clearly. The screenshot you see below is with the default values of DeNoise AI. The software did an outstanding job at reducing noise without me making any adjustments. Sharpness and detail were retained. I can also add more sharpness or recover details if needed.
When you’re satisfied with the image, you click Save Image. You’ll have options to choose the file format and color profile of the image file.
If you’re using DeNoise AI as a plugin, the rendered image with noise reduction will be reimported back to the host application.
The following image I tested was a photograph of the milky way. For this test, I invoked Topaz Studio 2 and used DeNoise AI as a plugin. With astrophotography, noise reduction can be quite difficult, especially with the trade-off of losing detail.
This time I made adjustments in DeNoise AI. I changed the processing model to AI Clear, set the Remove Noise to Medium, and Enhance Sharpness to High. The results were amazing. The software didn’t confuse stars for noise.
Topaz DeNoise vs Lightroom and Photoshop
The big question in this Topaz DeNoise AI review is how it holds up versus Photoshop and Lightroom.
The Adobe Camera RAW Engine is used in Lightroom to render RAW files and edit them. Photoshop also uses ACR for RAW files. When I compared the noise reduction in Photoshop, I used its Reduce Noise filter instead to test a different tool.
For the comparison, I used an image that was photographed on a cropped APS-C sensor Canon Rebel T5i camera. This camera doesn’t do well in low light. The camera settings were 1/10, f/5.6, and ISO 6400.
With Photoshop and Lightroom’s noise reduction tools, I had to play with the sliders significantly longer than with DeNoise AI.
Below is a comparison of the best results I could obtain with the different software.
Photoshop did a poor job at noise reduction. Lightroom reduced the noise, but too much detail was lost. DeNoise AI did the best job of reducing noise while retaining some detail. This was not an easy image for noise reduction software. But DeNoise AI did the best job at reducing noise.
Price
Topaz DeNoise regularly costs $79.99. You can also get a 30-day Topaz DeNoise trial.
You can also purchase the Topaz Labs Image Quality Bundle for $199.99, which saves $59.98. It includes:
- Topaz Denoise AI
- Topaz Gigapixel AI
- Topaz Sharpen AI
System Requirements
Requirements | Windows | Mac |
OS Version | 10, 11 64-bit only | macOS 10.14+ |
minimum display size | 1024 x 768 pixels | 1024 x 768 pixels |
OpenGL support level | Open GL 3.3 or later | Open GL 3.3 or later |
CPU | Intel | AMD |
minimum | Intel i5 or equivalent (3.0GHz and above) | Ryzen 5 or equivalent (3.0GHz and above) |
recommended | Intel i7 or greater (4GHz and above) | Ryzen 7 or greater (4GHz and above) |
GPU | Nvidia | AMD |
minimum | 2GB of dedicated VRAM (GT 740 or greater) | 2GB of dedicated VRAM (Radeon 5870 or greater) |
optimal | 6GB of dedicated VRAM (GTX 970 or greater) | 6GB of dedicated VRAM (Radeon RX 460 or greater) |
best | 16GB of dedicated VRAM (GTX 1080 or greater) | 16GB of dedicated VRAM (Radeon RX 580 or greater) |
DeNoise AI does not support Intel HD Graphics 4600 or lower.
A minimum RAM of 8 GB is required. 16 GB is recommended, and 32 GB is optimal.
An Internet connection is required to log into the software.
The Apple M1, M1 Pro, and M1 Max chips are also compatible. There is no information on M2 chips.
Verdict and Conclusion
We covered some essential details in this Topaz DeNoise AI review.
And I understand the regular price of $79.99 for noise reduction software is expensive for most people. If your main photography software produces acceptable results, then by all means, stick with it.
On the other hand, Topaz DeNoise AI is the best noise-reduction software I have used. It beats out Photoshop and Lightroom, hands down.
I’m also a consistent user of Capture One, and I can confirm DeNoise AI is better at noise reduction.
Photographers with cameras that shoot poorly in low light or mobile phone photographers will benefit the most from DeNoise AI. They can save money on their field equipment and reduce noise in posts.
Professional photographers will need to invest in the top software for noise reduction. Their livelihood depends on it. Hopefully, this Topaz DeNoise AI review helped you and removed your reservations.
Frequently Asked Questions
Is Topaz DeNoise worth it?
Yes, DeNoise is worth it. It’s the best noise-reduction software. Though it is costly for only a noise reduction tool, it can yield a high ROI if you sell large prints or display high-resolution photos online. It’s definitely worth it for professional photographers and you should read our full Topaz DeNoise AI review for more details.
Is Topaz DeNoise better than Lightroom?
Topaz DeNoise and Lightroom are two different software. But they do have one thing in common: they both do noise reduction. In this aspect, DeNoise is better at noise reduction than Lightroom. Remember, DeNoise is also available as a plugin for Lightroom. Read the full Topaz DeNoise AI review to see how it’s better than Lightroom.
Which is the best noise reduction software?
I believe Topaz DeNoise is the best noise-reduction software. Lightroom, Photoshop, Capture One, and many other image-editing software do well in noise reduction, but DeNoise is the best. Topaz DeNoise only focuses on noise reduction using AI. Go over our full Topaz DeNoise AI review to see the examples.
How does Topaz DeNoise work?
The simple answer is DeNoise uses machine learning (artificial intelligence) to determine what’s noise and detail in an image. It then only reduces actual noise and doesn’t adjust any detail. Take a look at the video and example photos in my Topaz DeNoise AI review to see for yourself.
I bought Topaz denoise AI a year ago and love the functionality. Unfortunately Topaz have now denied me using the software I paid for. I turned off updates because I read about this problem before buying it. Despite that they forced an upgrade on my system which made it unusable. Loading an image now takes 7 minutes, it was never fast (1 minute for a full Canon 5Ds image) but is now unusable. Topaz support will not revert back to the old version, so I now have no use for a program I paid for a year ago. Think twice before laying out money for this.
why would cmyk not be supported.. print matters. this should be front and center a feature. cmyk files so the printer doesn’t screw up
Hi Steven,
CMYK would be nice to have. I imagine CMYK is not supported because noise reduction is usually the first step in image editing, whereas color grading or soft proofing is the last step. And DeNoise isn’t meant for soft proofing. Just my thoughts.