Background Learnable Cascade for Zero-Shot Object Detection

TL;DR

This paper introduces Background Learnable Cascade (BLC), a novel framework that enhances zero-shot object detection by progressively refining visual-semantic alignment and learning background representations, leading to significant performance improvements.

Contribution

The paper proposes a multi-stage cascade structure, semantic information flow, and a learnable background region proposal network to improve ZSD performance.

Findings

BLC outperforms state-of-the-art methods on MS-COCO.

The cascade structure improves visual-semantic alignment.

Learnable background vectors reduce confusion between background and unseen classes.

Abstract

Zero-shot detection (ZSD) is crucial to large-scale object detection with the aim of simultaneously localizing and recognizing unseen objects. There remain several challenges for ZSD, including reducing the ambiguity between background and unseen objects as well as improving the alignment between visual and semantic concept. In this work, we propose a novel framework named Background Learnable Cascade (BLC) to improve ZSD performance. The major contributions for BLC are as follows: (i) we propose a multi-stage cascade structure named Cascade Semantic R-CNN to progressively refine the alignment between visual and semantic of ZSD; (ii) we develop the semantic information flow structure and directly add it between each stage in Cascade Semantic RCNN to further improve the semantic feature learning; (iii) we propose the background learnable region proposal network (BLRPN) to learn an appropriate word vector for background class and use this learned vector in Cascade Semantic R CNN, this design makes \Background Learnable" and reduces the confusion between background and unseen classes. Our extensive experiments show BLC obtains significantly performance improvements for MS-COCO over state-of-the-art methods.