Skip to yearly menu bar Skip to main content


Poster

Learning Deep Embeddings with Histogram Loss

Evgeniya Ustinova · Victor Lempitsky

Area 5+6+7+8 #130

Keywords: [ Deep Learning or Neural Networks ] [ (Application) Computer Vision ] [ Similarity and Distance Learning ]


Abstract:

We suggest a new loss for learning deep embeddings. The key characteristics of the new loss is the absence of tunable parameters and very good results obtained across a range of datasets and problems. The loss is computed by estimating two distribution of similarities for positive (matching) and negative (non-matching) point pairs, and then computing the probability of a positive pair to have a lower similarity score than a negative pair based on these probability estimates. We show that these operations can be performed in a simple and piecewise-differentiable manner using 1D histograms with soft assignment operations. This makes the proposed loss suitable for learning deep embeddings using stochastic optimization. The experiments reveal favourable results compared to recently proposed loss functions.

Live content is unavailable. Log in and register to view live content