Zero-Shot Object Detection by Hybrid Region Embedding

TL;DR

This paper introduces a novel zero-shot object detection method using hybrid region embeddings, addressing the challenge of detecting unseen classes without training data, and demonstrates promising results on custom datasets.

Contribution

The paper proposes a new zero-shot detection approach combining embeddings with detection frameworks and introduces custom datasets for evaluation.

Findings

The method achieves promising zero-shot detection performance.

Custom datasets facilitate evaluation of zero-shot detection methods.

Hybrid embeddings improve detection of unseen classes.

Abstract

Object detection is considered as one of the most challenging problems in computer vision, since it requires correct prediction of both classes and locations of objects in images. In this study, we define a more difficult scenario, namely zero-shot object detection (ZSD) where no visual training data is available for some of the target object classes. We present a novel approach to tackle this ZSD problem, where a convex combination of embeddings are used in conjunction with a detection framework. For evaluation of ZSD methods, we propose a simple dataset constructed from Fashion-MNIST images and also a custom zero-shot split for the Pascal VOC detection challenge. The experimental results suggest that our method yields promising results for ZSD.