Can liveness detection models generalise to synthetic media generation techniques they were never trained on? [D]

“`html

• A Reddit post asks whether liveness detection models can adapt to new synthetic media generation techniques that were not part of their training data.
• The core concern is about the temporal gap between model development and current deepfake generation quality, questioning if existing systems are robust against novel attacks.

“`

Stay ahead of AI. Get the most important stories delivered to your inbox — no spam, no noise.

Please enable JavaScript in your browser to complete this form.

Name Email

Name *

First

Last

Email *

Submit