Rethinking Annotation for Object Detection: Is Annotating Small-size Instances Worth Its Cost?

TL;DR

This paper investigates whether annotating small objects in images is cost-effective by testing detection methods that avoid small object annotations, finding that certain image scaling techniques can match baseline performance without costly annotations.

Contribution

It introduces and evaluates image upscaling and downscaling methods that reduce annotation costs while maintaining detection accuracy for small objects.

Findings

Upscaling at test time with domain gap remedies achieves comparable performance to fully annotated training.

Distilled single-path detector performs as well as baseline trained with complete annotations.

Rethinking annotation strategies can reduce costs without sacrificing detection quality.

Abstract

Detecting objects occupying only small areas in an image is difficult, even for humans. Therefore, annotating small-size object instances is hard and thus costly. This study questions common sense by asking the following: is annotating small-size instances worth its cost? We restate it as the following verifiable question: can we detect small-size instances with a detector trained using training data free of small-size instances? We evaluate a method that upscales input images at test time and a method that downscales images at training time. The experiments conducted using the COCO dataset show the following. The first method, together with a remedy to narrow the domain gap between training and test inputs, achieves at least comparable performance to the baseline detector trained using complete training data. Although the method needs to apply the same detector twice to an input image with different scaling, we show that its distillation yields a single-path detector that performs equally well to the same baseline detector. These results point to the necessity of rethinking the annotation of training data for object detection.