One of the most powerful and fascinating technologies of recent years is the harnessing of Artificial Intelligence, or AI, for a variety of graphics tasks that were previously impossible. You may remember the iconic scene in the film “Blade Runner,” where Harrison Ford enlarges a photograph and reveals detail which could not be there to get clues to solve the case. When the film came out that was impossible. Now 40 years later, it’s not only possible, it’s a product: AVCLabs Video Enhancer AI.
This is a sponsored article and was made possible by AVC. The actual contents and opinions are the sole views of the author who maintains editorial independence even when a post is sponsored.
Enlarge Video 400%
AVCLabs Video Enhancer AI is a software utility that takes low-resolution videos and enlarges them, augmenting the missing pixels intelligently to raise them to higher resolutions. The results are surprisingly good. The amount of controls you have over the process are minimal. Basically, you load the footage, preview a little bit if desired, then hit the big button and wait for your results.
If you have 480p videos and want to look at them on 4K TV, you can do it – up to a point. The algorithms in the AI’s brain will detect edges, smooth lines, ease jagged edges, identify human faces and even helpfully add eyes, noses and mouths if they are indistinct in the source footage. You can target a range of larger resolutions, like 720, 1080, 4K – even 8K if you’re enlarging HD imagery. Essentially, the result you get depends on the source material. It’s an easy-to-use powerful piece of software.
Science Not Magic
After seeing results from AVC Video Enhancer AI, you can be forgiven if you believed it was a magic bullet. It makes you feel as though you can feed it a dark, fluffy 240p YouTube image from 2006 and have it remake it as a sparkling 4K HD video indistinguishable from the real thing. Obviously, that’s only partially true. I know from my experience of using other AI image enhancement software that the quality of what you get is entirely dependent on what you put in.
AI image enhancement of the kind used in Video Enhancer is at its heart a kind of smart sharpener: it takes the images you throw at it and redraws the lines as it interprets them. It doesn’t just sharpen – it guesses intelligently where the line should fall in between those two shades or hues or areas on the screen. Where to draw the line is its whole deal. As a rule, this is an amazing effect, and fuzzy videos of quite low resolution can be upgraded to a much more watchable state for our massive modern screens.
As you can see, the result even with quite terrible video, is quite startling.
The only problems that arise take place when the AI has to do too much work. If the quality of the source video is too low, doesn’t have good contrast, is blurry, moving too fast and interlaced, the result is less like smooth video and more like stop-motion cutout animation. The uncanny valley effect kicks in, and although it looks like a much-improved version of the original video, there’s an uneasy feeling that the video is coming apart a little at the seams.
On the up side, if you have 720p video and you want to upgrade it to 1080p or 4K, the results are really impressive. If you want to upscale 4K to 8K, then it’s also very good. As a side point, you’re going to need some serious empty space on your hard drive for that. The quality and speed of the processing is smooth and tasteful. Using the controls, you can also adjust the brightness and contrast, saturation and even have a stab at removing that annoying interlace combing.
I did sometimes feel I wanted more control over the application with the smoothing routines, but in many ways, it’s a good idea to keep the users’ hands off the controls and let the computer handle it. You have just enough control – but not so much that you disturb the software’s aim.
Too much manual control would take away the point of the program. You need to let the AVC software do the grunt work, or it turns into a full time job for you. Sometimes the resulting image is extraordinary. The edges are sharpened and the flat area smoothed out, giving an overall impression of resolution the original image doesn’t have.
You do have to be careful using the face enhancement features. By chance, the video I found on my drive to test the software was an old capture of a laserdisc of Laurie Anderson in concert. The first part features her wearing a mask with dark holes for her eyes. The software helpfully added eyes and noses in some frames which didn’t have any, adding a surreal and slightly ooky quality to the output.
A really good feature is that you can pause a render and save it – which could take a day to process if it’s long – then automatically resume when you come back.
All in all, I liked the output quality of the software a lot, and as long as you keep your expectations realistic for what the software can do, you really won’t be disappointed.
AVCLabs Video Enhancer AI is an excellent picture-enhancement utility and produces terrific results with decent quality input. Currently, a subscription is $39.95 per month, but if you don’t want to subscribe, you can upgrade to a lifetime membership for a one-time payment of $299.90. Comparing the pricing to other top-of-the-line video enhancers, it’s comfortably cheaper.
Our latest tutorials delivered straight to your inbox