Abstract: Existing deepfake detectors predominantly process entire facial images as input, which limits their sensitivity to local forgery cues due to representation bias and information loss through ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results