Pixel Club: What's in a Face? Metric Learning for Face Characterization

Speaker:
Omry Sendik (Tel-Aviv University)
Date:
Tuesday, 3.12.2019, 10:00
Place:
Electrical Eng. Building 1061

We present a method for determining which facial parts (mouth, nose, etc.) best characterize an individual, given a set of that individual's portraits. We introduce a novel distinctiveness analysis of a set of portraits, which leverages the deep features extracted by a pre‐trained face recognition CNN and a hair segmentation FCN, in the context of a weakly supervised metric learning scheme. Our analysis enables the generation of a polarized class activation map (PCAM) for an individual's portrait via a transformation that localizes and amplifies the discriminative regions of the deep feature maps extracted by the aforementioned networks. A user study that we conducted shows that there is a surprisingly good agreement between the face parts that users indicate as characteristic and the face parts automatically selected by our method. We demonstrate a few applications of our method, including determining the most and the least representative portraits among a set of portraits of an individual, and the creation of facial hybrids: portraits that combine the characteristic recognizable facial features of two individuals. Our face characterization analysis is also effective for ranking portraits in order to find an individual's look‐alikes (Doppelgängers).

Bio:
Omry currently heads the SoC algorithms group (a total of roughly 40 algorithm engineers) in Samsung Israel R&D Center. His focus is on developing ISP, CV and ML algorithms, targeted to the automotive market (mainly ADAS). In addition to that, he's a PhD candidate in the school of Computer Science of Tel-Aviv University, under the joint supervision of Prof. Daniel Cohen-Or and Prof. Dani Lischinski. His main research interest include employing pre-trained neural networks for the purpose of image synthesis.

Omry completed his MSc in the school of Electrical Engineering of Tel-Aviv University, where he was supervised by Prof. Hagit Messer Yaron. His thesis was in the sampling theory realm. Prior to that, he obtained a BSc in Electrical Engineering and a BA in Physics both at the Technion.

Back to the index of events