Authors: Yu-Xiong Wang,Ross Girshick,Martial Hebert,Bharath Hariharan
Where published:
CVPR 2018 6
ArXiv: 1801.05401
Document:
PDF
DOI
Abstract URL: http://arxiv.org/abs/1801.05401v2
Humans can quickly learn new visual concepts, perhaps because they can easily
visualize or imagine what novel objects look like from different views.
Incorporating this ability to hallucinate novel instances of new concepts might
help machine vision systems perform better low-shot learning, i.e., learning
concepts from few examples. We present a novel approach to low-shot learning
that uses this idea. Our approach builds on recent progress in meta-learning
("learning to learn") by combining a meta-learner with a "hallucinator" that
produces additional training examples, and optimizing both models jointly. Our
hallucinator can be incorporated into a variety of meta-learners and provides
significant gains: up to a 6 point boost in classification accuracy when only a
single training example is available, yielding state-of-the-art performance on
the challenging ImageNet low-shot classification benchmark.