Rank of Experts: Detection Network Ensemble

TL;DR

This paper introduces Rank of Experts, a simple class-wise ensemble detection method that ranks detectors per class based on average precision, improving large-scale object detection without joint training of models.

Contribution

The paper proposes a novel class-wise ensemble detection method called Rank of Experts that ranks and combines detectors per class based on their performance.

Findings

Achieved 2nd place in ILSVRC 2017 object detection competition.

Effective ensemble method without joint training of detectors.

Improved detection performance for large-scale object detection.

Abstract

The recent advances of convolutional detectors show impressive performance improvement for large scale object detection. However, in general, the detection performance usually decreases as the object classes to be detected increases, and it is a practically challenging problem to train a dominant model for all classes due to the limitations of detection models and datasets. In most cases, therefore, there are distinct performance differences of the modern convolutional detectors for each object class detection. In this paper, in order to build an ensemble detector for large scale object detection, we present a conceptually simple but very effective class-wise ensemble detection which is named as Rank of Experts. We first decompose an intractable problem of finding the best detections for all object classes into small subproblems of finding the best ones for each object class. We then solve the detection problem by ranking detectors in order of the average precision rate for each class, and then aggregate the responses of the top ranked detectors (i.e. experts) for class-wise ensemble detection. The main benefit of our method is easy to implement and does not require any joint training of experts for ensemble. Based on the proposed Rank of Experts, we won the 2nd place in the ILSVRC 2017 object detection competition.