Improving Object Localization with Fitness NMS and Bounded IoU Loss

11/01/2017
by   Lachlan Tychsen-Smith, et al.
0

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. 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[0.5:0.95] of 33.6%@79Hz and 41.8%@5Hz for MSCOCO and a Titan X (Maxwell).

READ FULL TEXT
research
03/03/2023

Confidence-driven Bounding Box Localization for Small Object Detection

Despite advancements in generic object detection, there remains a perfor...
research
07/14/2023

MPDIoU: A Loss for Efficient and Accurate Bounding Box Regression

Bounding box regression (BBR) has been widely used in object detection a...
research
08/04/2016

UnitBox: An Advanced Object Detection Network

In present object detection systems, the deep convolutional neural netwo...
research
06/06/2019

Omnidirectional Scene Text Detection with Sequential-free Box Discretization

Scene text in the wild is commonly presented with high variant character...
research
09/23/2022

Transformer-Based Microbubble Localization

Ultrasound Localization Microscopy (ULM) is an emerging technique that e...
research
07/26/2018

A Better Baseline for AVA

We introduce a simple baseline for action localization on the AVA datase...
research
08/03/2022

Statistical Attention Localization (SAL): Methodology and Application to Object Classification

A statistical attention localization (SAL) method is proposed to facilit...

Please sign up or login with your details

Forgot password? Click here to reset