DBF: Dynamic Belief Fusion for Combining Multiple Object Detectors

TL;DR

The paper introduces Dynamic Belief Fusion (DBF), a score-level fusion method that combines multiple object detectors by estimating detection confidence and optimally fusing their outputs, significantly improving detection accuracy.

Contribution

The paper presents a novel fusion approach that directly integrates detection scores using confidence modeling and Dempster's rule, outperforming existing methods.

Findings

DBF achieves higher detection accuracy than baseline fusion methods.

DBF outperforms individual detectors on ARL, PASCAL VOC 07, and 12 datasets.

The approach effectively estimates ambiguity and fuses multiple detector outputs.

Abstract

In this paper, we propose a novel and highly practical score-level fusion approach called dynamic belief fusion (DBF) that directly integrates inference scores of individual detections from multiple object detection methods. To effectively integrate the individual outputs of multiple detectors, the level of ambiguity in each detection score is estimated using a confidence model built on a precision-recall relationship of the corresponding detector. For each detector output, DBF then calculates the probabilities of three hypotheses (target, non-target, and intermediate state (target or non-target)) based on the confidence level of the detection score conditioned on the prior confidence model of individual detectors, which is referred to as basic probability assignment. The probability distributions over three hypotheses of all the detectors are optimally fused via the Dempster's combination rule. Experiments on the ARL, PASCAL VOC 07, and 12 datasets show that the detection accuracy of the DBF is significantly higher than any of the baseline fusion approaches as well as individual detectors used for the fusion.