Erratum to: Refining deep convolutional features for improving fine-grained image recognition
Weixia Zhang, Jia Yan, Wenxuan Shi, Tianpeng Feng, Dexiang Deng · EURASIP Journal on Image and Video Processing · 2017
1], it was noticed that there were several blanks in the Table 5 and the footnote of the Table 5, 'The 'n/a' entries in the table means that bounding box or part annotation is not used.' was incorrectly given as 'The 'n/a' entries in the table means that the results are not available.'This has now been acknowledged and corrected in this erratum.This has now been incorporated in the new Table 5 shown below.