Live Face De-Identification in Video
This addresses privacy concerns for individuals in video surveillance or media by providing an efficient de-identification tool, though it appears incremental as it builds on existing encoder-decoder architectures.
The paper tackles the problem of face de-identification in video by proposing a method to automatically modify faces at high frame rates, achieving decorrelation of identity while preserving perception aspects like pose and expression, resulting in natural-looking sequences with minimal distortion.
We propose a method for face de-identification that enables fully automatic video modification at high frame rates. The goal is to maximally decorrelate the identity, while having the perception (pose, illumination and expression) fixed. We achieve this by a novel feed-forward encoder-decoder network architecture that is conditioned on the high-level representation of a person's facial image. The network is global, in the sense that it does not need to be retrained for a given video or for a given identity, and it creates natural looking image sequences with little distortion in time.