A method for image processing in video conferencing, for correcting the gaze of an interlocutor (9) in an image or a sequence of images captured by at least one real camera (1), comprises the steps of • the at least one real camera (1) acquiring an original image of the interlocutor (9); • synthesizing a corrected view of the interlocutor's (9) face as seen by a virtual camera (3), the virtual camera (3) being located on the interlocutor's (9) line of sight and oriented towards the interlocutor (9); • transferring the corrected view of the interlocutor's (9) face from the synthesized view into the original image (10), thereby generating a final image (12); • at least one of displaying the final image (12) and transmitting the final image (12).