A specialized face-processing network consistent with the representational geometry of monkey face patches by Amirhossein Farzmahdi, et al.
Abstract:
Ample evidence suggests that face processing in human and non-human primates is performed differently compared with other objects. Converging reports, both physiologically and psychophysically, indicate that faces are processed in specialized neural networks in the brain -i.e. face patches in monkeys and the fusiform face area (FFA) in humans. We are all expert face-processing agents, and able to identify very subtle differences within the category of faces, despite substantial visual and featural similarities. Identification is performed rapidly and accurately after viewing a whole face, while significantly drops if some of the face configurations (e.g. inversion, misalignment) are manipulated or if partial views of faces are shown due to occlusion. This refers to a hotly-debated, yet highly-supported concept, known as holistic face processing. We built a hierarchical computational model of face-processing based on evidence from recent neuronal and behavioural studies on faces processing in primates. Representational geometries of the last three layers of the model have characteristics similar to those observed in monkey face patches (posterior, middle and anterior patches). Furthermore, several face-processing-related phenomena reported in the literature automatically emerge as properties of this model. The representations are evolved through several computational layers, using biologically plausible learning rules. The model satisfies face inversion effect, composite face effect, other race effect, view and identity selectivity, and canonical face views. To our knowledge, no models have so far been proposed with this performance and agreement with biological data.
The article runs a full forty-eight (48) pages of citation laden text.
If you want a shorter synopsis, try: Human Face Recognition Found In Neural Network Based On Monkey Brains, which summarizes the paper and mentions the following similarities between human facial recognition and recognition by the neural network:
- Both recognize faces easiest when seen between a full frontal and a profile
- Both have difficulty recognizing faces when upside down
- Composite faces, top and bottom from different people, are recognized by both as different people
- If the neural network is trained on one race, has difficulty recognizing faces of other races, just like people
A large amount of investigation remains to be done, along with extending the methodology used here to explore and create the neural network.
From a privacy/security perspective, counter-measures will be needed to defeat ever more accurate facial recognition software.