Nothing ruins footage quite like grainy, unpleasant digital noise. We've already done a breakdown of what it is and what causes it here, but today we thought we'd take you through three easy ways to fix the problem.
Fixing image noise
The easiest way to avoid image noise is to make sure you get the best shot you possibly can. Here's why: Image noise consists of random variations of brightness or color information in the shot you are trying to capture. It is most likely to show up:
- In low-light conditions
- When shooting with small sensors.
The mechanics are pretty simple: the more you try to compensate for low-light settings with gain and ISO, the more noise is produced as a result. So the best way to avoid it is to make sure you are shooting with good equipment in good lighting.
But let's say you've already filmed the footage, only to review it later and find that there is an unacceptable level of digital noise in the output. Is there any way to remove it in post?
As it turns out, there are a few softwares that you can use to fix the issue. We tried three on some noisy footage we had to take you through the process of using each one.
Pixop's AI denoiser does all the heavy lifting.
First up is Pixop's own AI-powered denoiser. Here is a breakdown of its features and the denoising process:
Free (with restrictions) and Pay-per-use
Since Pixop is cloud-based, all you need is a web browser, an internet connection and a Pixop account.
Our denoiser can reduce noise in digital video in an automated fashion, as opposed to going through the time-consuming task of hand-tuning multiple parameters and/or noise profiles using off-the-shelf video editing packages and plugins. This AI filter can reduce:
- Gaussian video noise
- Minor scanline jittering
- Aliasing that results from e.g. (de-)interlacing
- Compression artifacts
Here's some of what we've focused on when developing our denoiser:
- We harness the power of deep neural network architecture to help you enhance your video
- Our denoiser works to reduce noise in footage of any resolution up to and including UHD 4K.
- Set it and forget it. Our denoiser makes denoising a video as simple as choosing a filter and clicking 'process'.
- Engineered to work well on a range of video qualities, from raw to very degraded, lossy compressed video.
- Tested to produce good results on a wide range of different genres such as entertainment, sports, action and documentaries.
How to use
Since Pixop is a cloud-based service, there are no expensive plugins to install, software requirements, or hardware setups to worry about. All you have to do to get started is sign up for an account. Once you've created your account, simply log in and follow these steps to denoise your video:
1. Upload the video you want to denoise
2. Once it's done uploading, click 'process'.
3. Choose 'Pixop Denoiser' under filters.
4. Click 'process'.
It's as simple as that. Our denoiser will take care of the rest, without you having to worry about manually tweaking anything.
We recently launched version 2.0 of our denoiser, where we integrated a new deep learning noise estimation module (to help retain fine details/textures and better reduce noise) and we've switched our architecture and filter training (to better deal with camera panning and object motion). Our new denoiser:
- Is able to handle camera panning and object motion much better than the old model
- Retains fine details and textures better in low-noise situations
- Is able to better reduce noise in cases of bad noise problems
Final Cut Pro
Here's a look at Final Cut Pro's Noise Reduction effect.
Next up, we have Apple's Final Cut Pro.
Flat, one-time price of $299.99
MacOS 10.15.6 or later, 4GB of RAM (8GB is recommended for 4K editing, 3D titles, and 360-degree video editing), and a Metal-capable graphics card (1GB VRAM is recommended for 4K, 3D title, and 360-degree editing).
Apple's Final Cut Pro is its professional and prosumer video editing software. Though not strictly a video enhancement software, it does contain a Noise reduction clip effect that you can apply to your footage to improve clips shot in low-light conditions or with small-sensor devices such as smartphones or action cameras.
- In-built effect without the need for third-party plugins
- Can control the level of noise reduction
- Can control sharpness level
How to use
1. Created a project
2. Imported the video clip you want to denoise
3. Dragged the clip to the timeline and
4. Selected the clip
5. Press Command + 5 to open the effects menu
6. Locate the 'noise reduction' effect under the 'Basic' category
7. Drag the effect on top of the clip
8. Adjust the levels of noise reduction and sharpness
We used Topaz Labs' free trial to create this, which is why you see a watermark.
And finally, let's take a look at Topaz Labs' Video Enhance AI.
One-time price of $299.99 (includes one year of updates)
Topaz Labs recommend that you run Video Enhance AI on high-end Nvidia or AMD dedicated graphics cards, with the minimum requirement being a GPU that is DirectX12 compatible with 2GB VRAM or similar. It is also compatible with the Apple M1.
For those without dedicated Nvidia or AMD graphics cards, the software can also run on Intel iGPU, though it will be 5 times slower. If you do not have an Intel iGPU, the software will use your CPU, which can be 10 times slower than processing on a dedicated GPU. The CPU must have AVX2 instruction set (2015 or newer for Intel, 2016 or newer for AMD).
Topaz Video Enhance AI also works on Macs. Apart from the requirements mentioned above, the following also apply: Mac OS 10.15 Catalina to run on GPU for all models except Chronos (Slow motion/FPS conversion) which requires OS 11 Big Sur. With Mac OS 10.14 Mojave and newer, you can run on CPU.
Topaz Labs' Video Enhance AI is an AI-based consumer desktop video upscaler, which allows you to upscale, denoise, restore and deinterlace video while resulting in fewer motion artifacts, recovering details and giving more natural results.
- Topaz Labs works directly with hardware manufacturers to ensure the software works seamlessly on a number of machines
- It features a simple interface and convenient workflow
- Take as much or as little control as you need to by processing with Topaz's suggested settings and AI models or choosing your own.
How to use
1. Open the software and import a video
2. Once the video is imported, it will automatically be analyzed and Topaz Labs will suggest settings and an AI model to use in its 'suggested' pane.
3. If you would like to change the settings or AI model, click on 'All' and make the necessary adjustments.
4. Click process.
Cloud-based versus On-premises
While there are a number of differences between the denoisers we tested above, such as the methods each uses to achieve the final denoised output and different ML models, we wanted to spend some time on what we think is one of the key differences.
At Pixop, we are big believers in the cloud and we're convinced it represents the future of video enhancement and upscaling. While on-premises software undoubtedly has its merits, we know that in offering our service via an intuitive cloud platform, we are making it accessible to as many people as possible.
Something that is easy to see from this article is that both Topaz Video Enhance AI and Final Cut Pro have an extensive list of software and hardware requirements that not everyone has access to or can afford to set up from scratch. In addition to that, video enhancement and upscaling are incredibly compute-intensive, making them inaccessible to a vast majority of video professionals and semi-professionals. On the other hand, all you need to get started with Pixop is a browser, an internet connection and an account. Once you're in, Pixop does all the heavy lifting for you because we've largely automated the process of video enhancement and upscaling, making it as simple as choosing a filter and clicking process.