Improving Object Localization with Fitness NMS and Bounded IoU Loss

TL;DR

This paper introduces Fitness NMS and a bounded IoU loss to enhance object detection accuracy and localization, demonstrating significant improvements on MSCOCO without sacrificing speed.

Contribution

It proposes a novel Fitness NMS method and a new IoU-based bounding box regression loss, advancing object detection precision and efficiency.

Findings

Improved MAP at higher localization accuracy without speed loss

Achieved 33.6% MAP at 79Hz and 41.8% at 5Hz on MSCOCO

Enhanced detection performance with source code available

Abstract

We demonstrate that many detection methods are designed to identify only a sufficently accurate bounding box, rather than the best available one. To address this issue we propose a simple and fast modification to the existing methods called Fitness NMS. This method is tested with the DeNet model and obtains a significantly improved MAP at greater localization accuracies without a loss in evaluation rate, and can be used in conjunction with Soft NMS for additional improvements. Next we derive a novel bounding box regression loss based on a set of IoU upper bounds that better matches the goal of IoU maximization while still providing good convergence properties. Following these novelties we investigate RoI clustering schemes for improving evaluation rates for the DeNet wide model variants and provide an analysis of localization performance at various input image dimensions. We obtain a MAP of 33.6%@79Hz and 41.8%@5Hz for MSCOCO and a Titan X (Maxwell). Source code available from: https://github.com/lachlants/denet