Rethinking Intersection Over Union for Small Object Detection in Few-Shot Regime

by   Pierre Le Jeune, et al.

In Few-Shot Object Detection (FSOD), detecting small objects is extremely difficult. The limited supervision cripples the localization capabilities of the models and a few pixels shift can dramatically reduce the Intersection over Union (IoU) between the ground truth and predicted boxes for small objects. To this end, we propose Scale-adaptive Intersection over Union (SIoU), a novel box similarity measure. SIoU changes with the objects' size, it is more lenient with small object shifts. We conducted a user study and SIoU better aligns than IoU with human judgment. Employing SIoU as an evaluation criterion helps to build more user-oriented models. SIoU can also be used as a loss function to prioritize small objects during training, outperforming existing loss functions. SIoU improves small object detection in the non-few-shot regime, but this setting is unrealistic in the industry as annotated detection datasets are often too expensive to acquire. Hence, our experiments mainly focus on the few-shot regime to demonstrate the superiority and versatility of SIoU loss. SIoU improves significantly FSOD performance on small objects in both natural (Pascal VOC and COCO datasets) and aerial images (DOTA and DIOR). In aerial imagery, small objects are critical and SIoU loss achieves new state-of-the-art FSOD on DOTA and DIOR.


page 18

page 19


Cooperating RPN's Improve Few-Shot Object Detection

Learning to detect an object in an image from very few training examples...

Hallucination Improves Few-Shot Object Detection

Learning to detect novel objects from few annotated examples is of great...

A Comparative Attention Framework for Better Few-Shot Object Detection on Aerial Images

Few-Shot Object Detection (FSOD) methods are mainly designed and evaluat...

Rotated Object Detection via Scale-invariant Mahalanobis Distance in Aerial Images

Rotated object detection in aerial images is a meaningful yet challengin...

Multiscale IoU: A Metric for Evaluation of Salient Object Detection with Fine Structures

General-purpose object-detection algorithms often dismiss the fine struc...

Distinctive Fire and Smoke Detection with Self-Similar

Deep learning based object detection is demonstrating a preponderance in...

Fast Convergence for Object Detection by Learning how to Combine Error Functions

In this paper, we introduce an innovative method to improve the converge...

Please sign up or login with your details

Forgot password? Click here to reset