XS-VID: An Extremely Small Video Object Detection Dataset

TL;DR

The paper introduces XS-VID, a new dataset for small video object detection focusing on extremely small objects, and proposes YOLOFT, a method that improves detection accuracy for these challenging cases.

Contribution

It provides the first extensive dataset for extremely small objects in video detection and develops YOLOFT, a novel detection method that enhances feature association and temporal integration.

Findings

Existing methods underperform on small object detection.

XS-VID covers diverse scenes and tiny objects, enriching the dataset resources.

YOLOFT significantly improves detection accuracy and stability.

Abstract

Small Video Object Detection (SVOD) is a crucial subfield in modern computer vision, essential for early object discovery and detection. However, existing SVOD datasets are scarce and suffer from issues such as insufficiently small objects, limited object categories, and lack of scene diversity, leading to unitary application scenarios for corresponding methods. To address this gap, we develop the XS-VID dataset, which comprises aerial data from various periods and scenes, and annotates eight major object categories. To further evaluate existing methods for detecting extremely small objects, XS-VID extensively collects three types of objects with smaller pixel areas: extremely small (\textit{es}, $0\sim12^2$), relatively small (\textit{rs}, $12^2\sim20^2$), and generally small (\textit{gs}, $20^2\sim32^2$). XS-VID offers unprecedented breadth and depth in covering and quantifying minuscule objects, significantly enriching the scene and object diversity in the dataset. Extensive validations on XS-VID and the publicly available VisDrone2019VID dataset show that existing methods struggle with small object detection and significantly underperform compared to general object detectors. Leveraging the strengths of previous methods and addressing their weaknesses, we propose YOLOFT, which enhances local feature associations and integrates temporal motion features, significantly improving the accuracy and stability of SVOD. Our datasets and benchmarks are available at \url{https://gjhhust.github.io/XS-VID/}.