Revisiting the Fisher vector for fine-grained classification

Philippe-Henri Gosselin (1, 2) , Naila Murray (3) , Hervé Jégou (1) , Florent Perronnin (3)

1 TEXMEX - Multimedia content-based indexing
2 MIDI - Multimedia Indexation and Data Integration
3 Xerox Research Centre Europe [Meylan]

This paper describes the joint submission of Inria and Xerox to their joint participation to the FGCOMP'2013 challenge. Although the proposed system follows most of the standard Fisher classification pipeline, we describe a few key features and good practices that significantly improve the accuracy when specifically considering fine-grain classification tasks. In particular, we consider the late fusion of two systems both based on Fisher vectors, but for which we choose drastically design choices that make them very complementary. Moreover, we propose a simple yet effective filtering strategy, which significantly boosts the performance for several class domains.

Domains: Computer Vision and Pattern Recognition [cs.CV]

Philippe-Henri Gosselin, Naila Murray, Hervé Jégou, Florent Perronnin. Revisiting the Fisher vector for fine-grained classification. Pattern Recognition Letters, 2014, 49, pp.92-98. ⟨hal-01056223⟩