arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2403.09433cs.CV
收藏

基于元提示表示与实例对比优化的开放词汇目标检测

Open-Vocabulary Object Detection with Meta Prompt Representation and Instance Contrastive Optimization

The Chinese University of Hong Kong(香港中文大学);Huawei(华为)

Zhao Wang, Aoxue Li, Fengwei Zhou, Zhenguo Li, Qi Dou

AI总结:

针对开放词汇目标检测中基类过拟合和依赖额外数据的问题,提出元提示与实例对比学习框架,在不使用复杂技术的情况下,在LVIS、COCO和Objects365上超越现有方法,显著提升新类别泛化能力。

摘要:

经典目标检测器无法检测到之前未遇到的新类别对象。针对这一问题,提出了开放词汇目标检测(OVOD),旨在检测候选类别列表中的对象。然而,当前的OVOD模型存在对基类过拟合、严重依赖大规模额外数据以及训练过程复杂的问题。为了克服这些问题,我们提出了一种新颖的框架,包含元提示(Meta prompt)和实例对比学习(Instance Contrastive learning,MIC)方案。首先,我们模拟一个新颖类别出现的场景,以帮助学习类别和背景提示的提示学习器泛化到新类别。其次,我们设计了一种实例级对比策略,以促进类内紧凑性和类间分离,这有助于检测器对新类别对象的泛化。在检测器训练过程中,我们未使用知识蒸馏、集成模型或额外训练数据,我们的MIC在LVIS上优于使用这些复杂技术训练的先前最先进方法。最重要的是,MIC在新类别上表现出强大的泛化能力,例如,与COCO和Objects365上的先前最先进方法相比,分别提高了+4.3%和+1.9%的AP。

英文摘要:

Classical object detectors are incapable of detecting novel class objects that are not encountered before. Regarding this issue, Open-Vocabulary Object Detection (OVOD) is proposed, which aims to detect the objects in the candidate class list. However, current OVOD models are suffering from overfitting on the base classes, heavily relying on the large-scale extra data, and complex training process. To overcome these issues, we propose a novel framework with Meta prompt and Instance Contrastive learning (MIC) schemes. Firstly, we simulate a novel-class-emerging scenario to help the prompt learner that learns class and background prompts generalize to novel classes. Secondly, we design an instance-level contrastive strategy to promote intra-class compactness and inter-class separation, which benefits generalization of the detector to novel class objects. Without using knowledge distillation, ensemble model or extra training data during detector training, our proposed MIC outperforms previous SOTA methods trained with these complex techniques on LVIS. Most importantly, MIC shows great generalization ability on novel classes, e.g., with $+4.3\%$ and $+1.9\% \ \mathrm{AP}$ improvement compared with previous SOTA on COCO and Objects365, respectively.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑