Channel-Wise Spatiotemporal Aggregation Technology for Face Video Forensics

Table 1

Basic information of datasets used in this work.

Dataset	Real/fake	Frames (k)	Size
FF++	1000/4000	509.9/2039.6	480p,720p,1080p
DFDC-P	1131/4113	88.4/1783.3	180p–2160p
Celeb-DF	890/5639	358.8/2116.8	Multiple