Sampling Matters in Deep Embedding LearningEI

    Cited by: 31|Bibtex|34|

    ICCV, pp. 2859-2867, 2017.

    Abstract:

    Deep embeddings answer one simple question: How similar are two images? Learning these embeddings is the bedrock of verification, zero-shot learning, and visual search. The most prominent approaches optimize a deep convolutional network with a suitable loss function, such as contrastive loss or triplet loss. While a rich line of work focu...More
    Your rating :
    0

     

    Tags
    Comments