Object Detection via a Multi-region and Semantic Segmentation-Aware CNN Model

TLDR

The CNN representation captures diverse discriminative appearance factors and localization sensitivity essential for accurate object localization. The study proposes an object detection system using a multi‑region CNN that incorporates semantic segmentation‑aware features. The system employs an iterative localization mechanism that alternates between scoring box proposals and refining locations with a deep CNN regression model. The approach achieves high localization accuracy, attaining 78.2 % mAP on PASCAL VOC2007 and 73.9 % on VOC2012, surpassing prior work.

Abstract

We propose an object detection system that relies on a multi-region deep convolutional neural network (CNN) that also encodes semantic segmentation-aware features. The resulting CNN-based representation aims at capturing a diverse set of discriminative appearance factors and exhibits localization sensitivity that is essential for accurate object localization. We exploit the above properties of our recognition module by integrating it on an iterative localization mechanism that alternates between scoring a box proposal and refining its location with a deep CNN regression model. Thanks to the efficient use of our modules, we detect objects with very high localization accuracy. On the detection challenges of PASCAL VOC2007 and PASCAL VOC2012 we achieve mAP of 78.2% and 73.9% correspondingly, surpassing any other published work by a significant margin.