∆-encoder: An effective sample synthesis method for

Eli Schwartz, Leonid Karlinsky, Joseph Shtok, Sivan Harary, Mattias Marder, Abhishek Kumar, Rogerio Feris, Raja Giryes, Alex M. Bronstein

Research output: Contribution to journalConference articlepeer-review

59 Scopus citations

Abstract

Learning to classify new categories based on just one or a few examples is a long-standing challenge in modern computer vision. In this work, we propose a simple yet effective method for few-shot (and one-shot) object recognition. Our approach is based on a modified auto-encoder, denoted ∆-encoder, that learns to synthesize new samples for an unseen category just by seeing few examples from it. The synthesized samples are then used to train a classifier. The proposed approach learns to both extract transferable intra-class deformations, or "deltas", between same-class pairs of training examples, and to apply those deltas to the few provided examples of a novel class (unseen during training) in order to efficiently synthesize samples from that new class. The proposed method improves the state-of-the-art of one-shot object-recognition and performs comparably in the few-shot case.

Original languageEnglish
Pages (from-to)2845-2855
Number of pages11
JournalAdvances in Neural Information Processing Systems
Volume2018-December
StatePublished - 2018
Event32nd Conference on Neural Information Processing Systems, NeurIPS 2018 - Montreal, Canada
Duration: 2 Dec 20188 Dec 2018

Funding

FundersFunder number
ERC-StG
Intelligence Advanced Research Projects ActivityD17PC00341
Intelligence Advanced Research Projects Activity
European Research Council

    Fingerprint

    Dive into the research topics of '∆-encoder: An effective sample synthesis method for'. Together they form a unique fingerprint.

    Cite this