Timezone: »

 
Poster
Learning a Metric Embedding for Face Recognition using the Multibatch Method
Oren Tadmor · Tal Rosenwein · Shai Shalev-Shwartz · Yonatan Wexler · Amnon Shashua

Mon Dec 05 09:00 AM -- 12:30 PM (PST) @ Area 5+6+7+8 #98
in Monday Posters »
This work is motivated by the engineering task of achieving a near state-of-the-art face recognition on a minimal computing budget running on an embedded system. Our main technical contribution centers around a novel training method, called Multibatch, for similarity learning, i.e., for the task of generating an invariant ``face signature'' through training pairs of ``same'' and ``not-same'' face images. The Multibatch method first generates signatures for a mini-batch of $k$ face images and then constructs an unbiased estimate of the full gradient by relying on all $k^2-k$ pairs from the mini-batch. We prove that the variance of the Multibatch estimator is bounded by $O(1/k^2)$, under some mild conditions. In contrast, the standard gradient estimator that relies on random $k/2$ pairs has a variance of order $1/k$. The smaller variance of the Multibatch estimator significantly speeds up the convergence rate of stochastic gradient descent. Using the Multibatch method we train a deep convolutional neural network that achieves an accuracy of $98.2\%$ on the LFW benchmark, while its prediction runtime takes only $30$msec on a single ARM Cortex A9 core. Furthermore, the entire training process took only 12 hours on a single Titan X GPU.
Keywords: Deep Learning or Neural Networks · (Application) Computer Vision · (Application) Object and Pattern Recognition
[ PDF] 

Author Information

Oren Tadmor (OrCam)
Tal Rosenwein (Orcam)
Shai Shalev-Shwartz (OrCam)
Yonatan Wexler (OrCam)
Amnon Shashua (OrCam)

More from the Same Authors