TensorPix

Do you have a version that I can install on my computer and run locally?

We currently don't support on premise software installation. Our framework utilizes Deep Neural Networks that are very heavy on the computer resources. Because of that, one should have a powerful (and expensive) GPU card in order to run our software in reasonable time. Without a proper GPU, processing times for just a minute of SD video can be tens of hours or more. Furthermore, local installation is usually limited to processing one video at the time, per GPU. This isn't efficient if you want to enhance a large video archive.

How does your tool compare to Topaz Labs?

The biggest difference is that Topaz Labs is an offline software that you install on your computer, while TensorPix can be used through web browser. Because Topaz runs on your PC, it uses your hardware to process files. This becomes problematic if your PC isn't quite high-end (and most users don't have high-end computers). You can still use Topaz with a less powerful computer, but the processing speed will be heavily impaired, meaning you could wait for tens of hours for enhancing a mere couple of seconds of video. On the other side, TensorPix sends all processing to cloud which has access to

How much of a quality improvement can I expect?

The rule of thumb is - the better the quality aspect you're trying to fix in the source, the better the enhancement. E.g. if the source has severe compresssion artifacts that obscure a lot of details, it's hard to expect much improvement. However, if the source only has moderate or low artifacting, the end result would be much better as it's easier to fix minor artifacts than heavy ones. The best thing is try enhancing a sample and see how it turns out (this is why we have Free Previews for videos).

How fast is the enhancing process?

While the processing speed largely depends on the input file parameters such as resolution, framerate, and duration, you can expect the following: A minute of SD video will take 5-10 minutes to enhance. A single 2000x2000 image will process within few seconds

Do you claim any copyrights on the uploaded or enhanced videos or images?

We don't claim any copyrights on uploaded or on enhanced videos and images. You can delete your files at any time. When you delete a file, we completely delete it from our servers.

Is my data safe and who can see my data?

We have implemented rigorous technical and organizational security policies to protect your data. This includes, but it's not limited to: Using military grade 256-bit AES encryption algorithm for all data stored on our servers Using long and random passwords for all access points Using 2 Factor Authentication where applicable Hashing all user account passwords Regularly updating all our systems in order to avoid software exploits and hacks Always giving users an option to delete ALL their data Automatically deleting any unnecessary data such as temporary files Restricting user data access to m

What Makes AI Enhancement Better Than Regular Video Editing?

AI enhancement is automatic and smart - you don't need to learn complicated editing software or spend hours adjusting settings. Traditional video editing requires technical knowledge and lots of manual work. AI enhancement uses intelligent algorithms that know exactly what improvements to make based on millions of examples. Just upload and wait - the AI handles everything automatically and usually produces better results than manual editing, especially for quality improvements.

Do you train your AI models on user data?

No, our policy is not to use user data for training our AI models. We only use publicly available datasets and our own data to train the models. This means that your videos and images are not used to improve our AI, ensuring your privacy and security.