Scalable Object Detection using Deep Neural Networks

TL;DR

This paper introduces a scalable object detection method using deep neural networks that predicts multiple bounding boxes and scores, effectively handling multiple object instances and achieving competitive results on standard benchmarks.

Contribution

The work presents a saliency-inspired neural network model that predicts class-agnostic bounding boxes, enabling detection of multiple objects and cross-class generalization.

Findings

Achieved competitive detection performance on VOC2007 and ILSVRC2012 datasets.

Effectively handles multiple object instances with fewer neural network evaluations.

Predicts variable numbers of object locations per image.

Abstract

Deep convolutional neural networks have recently achieved state-of-the-art performance on a number of image recognition benchmarks, including the ImageNet Large-Scale Visual Recognition Challenge (ILSVRC-2012). The winning model on the localization sub-task was a network that predicts a single bounding box and a confidence score for each object category in the image. Such a model captures the whole-image context around the objects but cannot handle multiple instances of the same object in the image without naively replicating the number of outputs for each instance. In this work, we propose a saliency-inspired neural network model for detection, which predicts a set of class-agnostic bounding boxes along with a single score for each box, corresponding to its likelihood of containing any object of interest. The model naturally handles a variable number of instances for each class and allows for cross-class generalization at the highest levels of the network. We are able to obtain competitive recognition performance on VOC2007 and ILSVRC2012, while using only the top few predicted locations in each image and a small number of neural network evaluations.