FAST OBJECT LOCALIZATION VIA SENSITIVITY ANALYSIS

ICLR 2019  ·  Mohammad K. Ebrahimpour, David C. Noelle ·

Deep Convolutional Neural Networks (CNNs) have been repeatedly shown to perform well on image classification tasks, successfully recognizing a broad array of objects when given sufficient training data. Methods for object localization, however, are still in need of substantial improvement. Common approaches to this problem involve the use of a sliding window, sometimes at multiple scales, providing input to a deep CNN trained to classify the contents of the window. In general, these approaches are time consuming, requiring many classification calculations. In this paper, we offer a fundamentally different approach to the localization of recognized objects in images. Our method is predicated on the idea that a deep CNN capable of recognizing an object must implicitly contain knowledge about object location in its connection weights. We provide a simple method to interpret classifier weights in the context of individual classified images. This method involves the calculation of the derivative of network generated activation patterns, such as the activation of output class label units, with regard to each in- put pixel, performing a sensitivity analysis that identifies the pixels that, in a local sense, have the greatest influence on internal representations and object recognition. These derivatives can be efficiently computed using a single backward pass through the deep CNN classifier, producing a sensitivity map of the image. We demonstrate that a simple linear mapping can be learned from sensitivity maps to bounding box coordinates, localizing the recognized object. Our experimental results, using real-world data sets for which ground truth localization information is known, reveal competitive accuracy from our fast technique.

PDF Abstract

Datasets


Results from the Paper


  Submit results from this paper to get state-of-the-art GitHub badges and help the community compare results to other papers.

Methods


No methods listed for this paper. Add relevant methods here