Aiarty Video Enhancer Introduced – AI Tools for Cleaning Up Imperfect Footage

AISponsored

by Alexandra ThompsonJuly 17^th, 2025

Subscribe to News - Software

Aiarty, a new division of Digiarty Software (best known for their video conversion tools), has developed an AI-powered desktop application for filmmakers working with less-than-ideal footage. The Aiarty Video Enhancer is designed to help clean up noisy low-light scenes, soft focus, compression artifacts, old archival clips, and more. And it runs entirely offline – no cloud upload required. Let’s have a look at this new software!

Digiarty Software, founded in 2006 and based in Chengdu, China, has nearly two decades of experience in multimedia tools with a reported user base spanning over 180 countries. With Aiarty, they shifted into AI-powered imaging, focusing just on enhancement without the overhead of a full editing suite.

Key features of Aiarty Video Enhancer

• AI Denoise and Deblur: Reduces compression artifacts and motion blur using trained deep-learning models.
• Upscaling: Increases resolution up to 4K with options for 2×, 3×, or 4× upscaling.
• Slow Motion with Frame Interpolation: Creates smoother slow motion or motion-enhanced playback, up to 120fps.
• Audio Denoise: An Audio Enhancer removes background hiss and ambient noise (wind, environment) and enhances vocals.
• Batch Processing & GPU Acceleration: Supports multiple file workflows with hardware-based speedups (with supported graphics cards).
• Model Switching: Choose from three dedicated AI engines based on footage type and needs.

AI enhancement in post-production

Aiarty centers its workflow around three general-purpose models trained, according to Aiarty, on over 2.3 million real-world videos:

• moDetail-HQ v2 (Diffusion + GAN): Restores fine textures in faces, hair, and natural elements, effective for footage where clarity and detail are essential.
• Smooth-HQ v2 (Diffusion): Gently removes noise while preserving natural skin tones and colors. Designed for everyday video and casual environments.
• superVideo vHQ (Diffusion + GAN): Targets extreme low-light conditions like night skies and dim interiors where denoising and detail recovery are critical. This model processes more slowly but delivers strong results in challenging scenes.

Under-the-hood optimization

Aiarty is optimized to run efficiently on NVIDIA, AMD, and Intel GPUs, with particular tuning for NVIDIA’s RTX 30, 40, and new 50-series cards powered by Blackwell architecture. Under the hood, it uses TensorRT 10.9 and CUDA’s asynchronous processing for parallel frame rendering, along with smart model pruning and precision adjustments to reduce computational load.

An asynchronous pipeline replaces the typical serial frame-by-frame bottleneck, keeping GPU usage high and consistent. These choices should result in smoother performance, faster processing, and better energy efficiency, especially when working with high-resolution or longer clips.

Real-world performance and workflow

Aiarty uses GPU acceleration and a dedicated Turbo Mode to improve processing speed. Performance depends on clip length and resolution, but the software is designed to support both short videos and longer footage. Batch processing is supported, and the interface also lends itself to quick individual enhancements.

Price and availability

A free trial version is available with full access to features and no watermarks, so users can evaluate the software in their own workflow. A standard license is currently priced at $79 (billed annually) for one computer (regularly $99). And for those interested in a long-term option, don’t miss the 30% discount on a lifetime license ($165, down from $235) for use on up to three devices. This includes ongoing updates and does not require a subscription.

For more information on the Video Enhancer, please visit their website.

Would Aiarty find a place in your restoration workflow? Or do you prefer to tackle noise and resolution another way? Let us know in the comments below!