The KAI Catch app, which means that KAIST catches it, is a special forgery detection app. It is an app made by students in the Multimedia Lab of KAIST's Department of Computer Science and Technology, and provides deep fake and photo image forgery detection. If you upload a photo image, you can check whether the image has been transformed into a deep fake or a forged photo image.
The deep fake detection rate is around 95% for the same transformation as learning, and the technology reliability is over 90%, and in the case of photo image forgery detection, the detection rate is 90-95% and the technology reliability is at the level of 90-95%. In the case of photographic images, the reliability exceeds 95% when it is detected as forgery in both analyzed images. You can also request further in-depth analysis by email.
If it is difficult to interpret the analysis results in the Kai Catch app, or if you want additional detailed analysis, you can request an email analysis. Analysis result report can also be issued: Email address heunglee@kaist.ac.kr
However, separate analysis costs are incurred for email (additional) analysis requests, etc.
KAICATCH APP detects both the deepfake and photo image forgery. Deepfake detection rate is around 95% and the detection reliability is above 90%. In case of photo image forgery, the detection ratio is between 90 and 95%, and detection reliability is between 90% and 95%.
The deep fake detection rate is around 95% for the same transformation as learning, and the technology reliability is over 90%, and in the case of photo image forgery detection, the detection rate is 90-95% and the technology reliability is at the level of 90-95%. In the case of photographic images, the reliability exceeds 95% when it is detected as forgery in both analyzed images. You can also request further in-depth analysis by email.
If it is difficult to interpret the analysis results in the Kai Catch app, or if you want additional detailed analysis, you can request an email analysis. Analysis result report can also be issued: Email address heunglee@kaist.ac.kr
However, separate analysis costs are incurred for email (additional) analysis requests, etc.
KAICATCH APP detects both the deepfake and photo image forgery. Deepfake detection rate is around 95% and the detection reliability is above 90%. In case of photo image forgery, the detection ratio is between 90 and 95%, and detection reliability is between 90% and 95%.
Show More
