End-to-end training of object class detectors for mean average precision

Henderson, P. and Ferrari, V. (2017) End-to-end training of object class detectors for mean average precision. In: Lai, S.-H., Lepetit, V., Nishino, K. and Sato, Y. (eds.) Computer Vision – ACCV 2016: 13th Asian Conference on Computer Vision, Taipei, Taiwan, November 20-24, 2016, Revised Selected Papers, Part V. Series: Lecture notes in computer science (10115). Springer: Cham, pp. 198-213. ISBN 9783319541921 (doi: 10.1007/978-3-319-54193-8_13)

Full text not currently available from Enlighten.


We present a method for training CNN-based object class detectors directly using mean average precision (mAP) as the training loss, in a truly end-to-end fashion that includes non-maximum suppresion (NMS) at training time. This contrasts with the traditional approach of training a CNN for a window classification loss, then applying NMS only at test time, when mAP is used as the evaluation metric in place of classification accuracy. However, mAP following NMS forms a piecewise-constant structured loss over thousands of windows, with gradients that do not convey useful information for gradient descent. Hence, we define new, general gradient-like quantities for piecewise constant functions, which have wide applicability. We describe how to calculate these efficiently for mAP following NMS, enabling to train a detector based on Fast R-CNN [1] directly for mAP. This model achieves equivalent performance to the standard Fast R-CNN on the PASCAL VOC 2007 and 2012 datasets, while being conceptually more appealing as the very same model and loss are used at both training and test time.

Item Type:Book Sections
Glasgow Author(s) Enlighten ID:Henderson, Dr Paul
Authors: Henderson, P., and Ferrari, V.
College/School:College of Science and Engineering > School of Computing Science
Published Online:11 March 2017

University Staff: Request a correction | Enlighten Editors: Update this record