UnitBox: An Advanced Object Detection Network

TL;DR

UnitBox introduces an IoU-based loss function and a fully convolutional network architecture for improved object detection, achieving state-of-the-art results in face detection with fast convergence and robustness to object scale and shape variations.

Contribution

The paper proposes a novel IoU loss function integrated into a deep fully convolutional network for more accurate bounding box regression in object detection.

Findings

Achieved top performance on FDDB face detection benchmark.

Demonstrated robustness to object scale and shape variations.

Converges faster than previous methods.

Abstract

In present object detection systems, the deep convolutional neural networks (CNNs) are utilized to predict bounding boxes of object candidates, and have gained performance advantages over the traditional region proposal methods. However, existing deep CNN methods assume the object bounds to be four independent variables, which could be regressed by the $\ell_2$ loss separately. Such an oversimplified assumption is contrary to the well-received observation, that those variables are correlated, resulting to less accurate localization. To address the issue, we firstly introduce a novel Intersection over Union ($IoU$) loss function for bounding box prediction, which regresses the four bounds of a predicted box as a whole unit. By taking the advantages of $IoU$ loss and deep fully convolutional networks, the UnitBox is introduced, which performs accurate and efficient localization, shows robust to objects of varied shapes and scales, and converges fast. We apply UnitBox on face detection task and achieve the best performance among all published methods on the FDDB benchmark.