Your browser is out-of-date!

Update your browser to view this website correctly. Update my browser now

×

iSize’s new solution eliminates compression artifacts in user-generated content

BitClear works with content that has been through multiple transcoding iterations, reviving it to the maximum possible quality

iSize is aiming to make “unwatchable” videos watchable with AI-based video processing technology that removes compression artifacts (like blurring and blocking artifacts) from user-generated (or heavily-compressed) content.

BitClear works with content that has been through multiple transcoding iterations, reviving it to the maximum possible quality without impacting the artistic intent of the original creators. The process can also allow for video upscaling, all with as little as 5ms processing latency on GPUs or high-performance CPUs, said the company.

“The huge traffic in shared video content today passes through several platforms and users before potentially reaching a large audience or big influencers,” explained Sergio Grce, CEO and founder of iSIZE.

“Quite often, user-generated content (UGC) that is viewed by millions of people over multiple social media platforms has been re-shared and re-uploaded numerous times. This process tends to make it very degraded or even unwatchable due to multiple transcoding iterations.

“For social media or UGC distribution and streaming companies, it is imperative that video content is presented in as high quality as possible to retain audiences,” added Grce. “BitClear is designed to achieve the maximum quality improvement by alleviating the effect of compression artifacts. Its AI nature and implementation efficiency means it can be deployed at scale without the need for human inspection or tuning.”

The solution uses a bespoke neural network to disentangle the noise from the data manifolds, it is then able to remove compression noise and retain or recover the original content features, said iSize. BitClear learns the noise signatures of the various encoding standards, without needing to know the history of the specific asset, meaning it can process any highly-compressed content and produce a higher-quality output that improves the value of the asset.