What Is Holding Back Convnets for Detection

Authors: Bojan Pepik Rodrigo Benenson Tobias Ritschel Bernt Schiele

Publish Date: 2015/10/07

Volume: , Issue: , Pages: 517-528

PDF Link

Abstract

Convolutional neural networks have recently shown excellent results in general object detection and many other tasks Albeit very effective they involve many userdefined design choices In this paper we want to better understand these choices by inspecting two key aspects “what did the network learn” and “what can the network learn” We exploit new annotations Pascal3D+ to enable a new empirical analysis of the RCNN detector Despite common belief our results indicate that existing stateoftheart convnets are not invariant to various appearance factors In fact all considered networks have similar weak points which cannot be mitigated by simply increasing the training data architectural changes are needed We show that overall performance can improve when using image renderings as data augmentation We report the best known results on Pascal3D+ detection and viewpoint estimation tasksOpen Access This chapter is licensed under the terms of the Creative Commons AttributionNonCommercial 25 International License http//creativecommonsorg/licenses/bync/25/ which permits any noncommercial use sharing adaptation distribution and reproduction in any medium or format as long as you give appropriate credit to the original authors and the source provide a link to the Creative Commons license and indicate if changes were madeThe images or other third party material in this chapter are included in the chapters Creative Commons license unless indicated otherwise in a credit line to the material If material is not included in the chapters Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use you will need to obtain permission directly from the copyright holder

Keywords: