公开(公告)号：EP4449382A1

公开(公告)日：2024-10-23

申请号：EP22834635.9

申请日：2022-12-13

申请人： THALES DIS FRANCE SAS ,  THALES ,  Board Of Trustees Of Michigan State University ,  INRIA - Institut National de Recherche en Informatique et en Automatique

发明人： ANGHELONE, David ,  FAURE, Philippe ,  CHEN, Cunjian ,  ROSS, Arun ,  DANTCHEVA, Antitza

IPC分类号： G06V40/70 ,  G06V10/82 ,  G06V40/16

公开(公告)号：EP4198928A1

公开(公告)日：2023-06-21

申请号：EP21306772.1

申请日：2021-12-14

申请人： Thales Dis France SAS ,  THALES ,  Board of Trustees of Michigan State University ,  INRIA - Institut National de Recherche en Informatique et en Automatique

发明人： ANGHELONE, David ,  FAURE, Philippe ,  CHEN, Cunjian ,  ROSS, Arun ,  DANTCHEVA, Antitza

IPC分类号： G06V40/70 ,  G06V10/82 ,  G06V40/16

摘要： A cross-spectral face recognition learning method according to the invention is based on a set of associated face images, a thermal image (10T) and a visual image (10V), of a plurality of persons. The thermal image (10T) is coded in two different ways. A style encoder (200T) provides a style code (420T) of the thermal image (10T). An identity encoder (100T) provides an identity code (410T) of the thermal image (10T). The visual image 10V is coded in a similar way with a style encoder (200V) providing a style code (420V) and with an identity encoder (100V) providing an identity code (410V). The two face images of the same person share in the identity features a common part in the respective identity codes (410T and 410V), noted as common identity code (410VT), whereas the style codes (420T and 420V) for the two images comprise features only relevant two the specific style, i.e. either thermal or visual, of the image.