Ensembling object detectors for image and video data analysis

TL;DR

This paper introduces an ensembling method for object detectors that enhances detection accuracy and bounding box precision in images and videos, with applications in annotation and tracking.

Contribution

It presents a novel ensembling approach for image and video object detection, including a tracking-based refinement scheme for videos.

Findings

Improved detection performance through ensembling.

Enhanced bounding box precision in images and videos.

Effective as a standalone detection or annotation framework.

Abstract

In this paper, we propose a method for ensembling the outputs of multiple object detectors for improving detection performance and precision of bounding boxes on image data. We further extend it to video data by proposing a two-stage tracking-based scheme for detection refinement. The proposed method can be used as a standalone approach for improving object detection performance, or as a part of a framework for faster bounding box annotation in unseen datasets, assuming that the objects of interest are those present in some common public datasets.