Skip to main content
placeholder image

In defense of soft-assignment coding

Conference Paper


Abstract


  • In object recognition, soft-assignment coding enjoys

    computational efficiency and conceptual simplicity. How-

    ever, its classification performance is inferior to the newly

    developed sparse or local coding schemes. It would be

    highly desirable if its classification performance could be-

    come comparable to the state-of-the-art, leading to a coding

    scheme which perfectly combines computational efficiency

    and classification performance. To achieve this, we revisit

    soft-assignment coding from two key aspects: classification

    performance and probabilistic interpretation. For the first

    aspect, we argue that the inferiority of soft-assignment cod-

    ing is due to its neglect of the underlying manifold structure

    of local features. To remedy this, we propose a simple mod-

    ification to localize the soft-assignment coding, which sur-

    prisingly achieves comparable or even better performance

    than existing sparse or local coding schemes while main-

    taining its computational advantage. For the second as-

    pect, based on our probabilistic interpretation of the soft-

    assignment coding, we give a probabilistic explanation to

    the magic max-pooling operation, which has successfully

    been used by sparse or local coding schemes but still poorly

    understood. This probability explanation motivates us to

    develop a new mix-order max-pooling operation which fur-

    ther improves the classification performance of the pro-

    posed coding scheme. As experimentally demonstrated, the

    localized soft-assignment coding achieves the state-of-the-

    art classification performance with the highest computa-

    tional efficiency among the existing coding schemes.

Authors


  •   Liu, Lingqiao (external author)
  •   Wang, Lei
  •   Liu, Xinwang (external author)

Publication Date


  • 2011

Citation


  • Liu, L., Wang, L. & Liu, X. (2011). In defense of soft-assignment coding. 2011 IEEE International Conference on Computer Vision (pp. 2486-2493). USA: IEEE.

Scopus Eid


  • 2-s2.0-84863044549

Ro Metadata Url


  • http://ro.uow.edu.au/infopapers/3809

Has Global Citation Frequency


Start Page


  • 2486

End Page


  • 2493

Place Of Publication


  • USA

Abstract


  • In object recognition, soft-assignment coding enjoys

    computational efficiency and conceptual simplicity. How-

    ever, its classification performance is inferior to the newly

    developed sparse or local coding schemes. It would be

    highly desirable if its classification performance could be-

    come comparable to the state-of-the-art, leading to a coding

    scheme which perfectly combines computational efficiency

    and classification performance. To achieve this, we revisit

    soft-assignment coding from two key aspects: classification

    performance and probabilistic interpretation. For the first

    aspect, we argue that the inferiority of soft-assignment cod-

    ing is due to its neglect of the underlying manifold structure

    of local features. To remedy this, we propose a simple mod-

    ification to localize the soft-assignment coding, which sur-

    prisingly achieves comparable or even better performance

    than existing sparse or local coding schemes while main-

    taining its computational advantage. For the second as-

    pect, based on our probabilistic interpretation of the soft-

    assignment coding, we give a probabilistic explanation to

    the magic max-pooling operation, which has successfully

    been used by sparse or local coding schemes but still poorly

    understood. This probability explanation motivates us to

    develop a new mix-order max-pooling operation which fur-

    ther improves the classification performance of the pro-

    posed coding scheme. As experimentally demonstrated, the

    localized soft-assignment coding achieves the state-of-the-

    art classification performance with the highest computa-

    tional efficiency among the existing coding schemes.

Authors


  •   Liu, Lingqiao (external author)
  •   Wang, Lei
  •   Liu, Xinwang (external author)

Publication Date


  • 2011

Citation


  • Liu, L., Wang, L. & Liu, X. (2011). In defense of soft-assignment coding. 2011 IEEE International Conference on Computer Vision (pp. 2486-2493). USA: IEEE.

Scopus Eid


  • 2-s2.0-84863044549

Ro Metadata Url


  • http://ro.uow.edu.au/infopapers/3809

Has Global Citation Frequency


Start Page


  • 2486

End Page


  • 2493

Place Of Publication


  • USA