How can I use your product?
We provide Deepware Scanner in web-platform, API, and SDK forms. You can either use our website, our API key, or in an offline environment via SDK. Contact us to learn more.
Do you detect any fake videos?
No, we are only dedicated to detecting AI-generated face manipulations.
What does the NO DEEPFAKE DETECTED(50%) result mean?
The model is looking for signs of manipulation on the human face. If there is no face on the video, you get this result.
What is the maximum video length we can scan?
So far, we have a maximum limit of 10 minutes for each video.
What are the platforms visitors can scan video from?
We currently scan videos from platforms such as YouTube, Facebook, and Twitter.
Can I trust your result 100%?
Deepfakes are not a solved problem yet. So our results are stating the chances of a specific video can be whether deepfake or not.
Do you have a mobile app?
We have an Android app that you can download right now. We are also working on the iOS app.
What are the things Deepware Scanner focuses on while scanning?
Deepware Scanner focuses on deepfake manipulation that uses a real video of a person and modifies or swaps their face with another person. We do not detect whether the voice is manipulated but only the face of a person in the video. Please check our GitHub page to read more about Deepware Scanner features.
What is the technical explanation of the video scanning process?
Empirical analysis has shown that 1 FPS is a good compromise between scanning speed and detection accuracy. Clustering faces allows us to remove noise such as falsely detected faces. Only the consistently detected faces will be clustered, and the detection score will be calculated only for them.
What is your detection accuracy?
Please visit our GitHub page to learn more about our detection accuracy with detailed information.
What are the advantages of API when you compare it with a web-platform?
API allows you to scan numerous videos at once. It can be integrated into platforms.
Do you scan and detect all types of synthetic media files?
No, we currently scan and detect only video files. We are working on detecting synthetically generated voice media files. However, we do not plan to detect synthetically generated images and text in our model.
What is the best video resolution to scan to get the best Deepware Scanner detection result?
We recommend scanning videos that are at least 1920 x 1080.
Does Deepware create deepfake videos as well?
Yes, as Deepware, we have a team on both sides to detect and generate deepfakes. We believe that to achieve the best detection rate, we also need to create deepfake videos. As we create better deepfake videos, our detection model gets better. Please contact us to learn more about our generation team. Check out our YouTube channel to see the videos we’ve generated so far.