超越对比的一致性：通过上下文一致性学习增强开放词表目标检测鲁棒性
Consistency Beyond Contrast: Enhancing Open-Vocabulary Object Detection Robustness via Contextual Consistency Learning
提出上下文一致性学习框架，通过生成同一物体在不同背景下的图像并施加特征一致性损失，提升开放词表检测器对背景变化的鲁棒性，可直接用于增强现有开放词表检测模型的泛化能力。
原文依据 Contextual Bootstrapped Data Generation (CBDG) and Contextual Consistency Loss (CCLoss)
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论近期开放词表目标检测的进展主要集中于两个方面：扩大数据集规模以及利用对比学习对齐语言与视觉模态。然而，这些方法往往忽略了单一模态内部的 consistency，尤其是当背景或环境发生变化时。这种一致性的缺失导致性能下降，因为模型难以在不同场景中检测同一物体，这揭示了一个鲁棒性差距。为解决该问题，我们引入上下文一致性学习（CCL），一种新颖的框架，集成了两个关键策略：上下文自举数据生成（CBDG）与上下文一致性损失（CCLoss）。CBDG 作为一种数据生成机制，生成包含相同物体但背景多样的图像。这是必要的，因为现有数据集本身无法支持我们的 CCL 框架。CCLoss 进一步强制物体特征在环境变化下保持不变，从而提升模型在不同场景中的鲁棒性。这些策略共同构成一个统一框架，用于确保同一模态内的上下文一致性。我们的方法取得了最优性能，在 OmniLabel 上超过先前方法 +16.3 AP，在 D^3 上超过 +14.9 AP。这些结果证明了强制模态内一致性的重要性，显著增强了模型在多样环境中的泛化能力。我们的代码已公开于：https://github.com/bozhao-li/CCL
英文原文摘要 1528 字符
Recent advances in open-vocabulary object detection focus primarily on two aspects: scaling up datasets and leveraging contrastive learning to align language and vision modalities. However, these approaches often neglect internal consistency within a single modality, particularly when background or environmental changes occur. This lack of consistency leads to a performance drop because the model struggles to detect the same object in different scenes, which reveals a robustness gap. To address this issue, we introduce Contextual Consistency Learning (CCL), a novel framework that integrates two key strategies: Contextual Bootstrapped Data Generation (CBDG) and Contextual Consistency Loss (CCLoss). CBDG functions as a data generation mechanism, producing images that contain the same objects across diverse backgrounds. This is essential because existing datasets alone do not support our CCL framework. The CCLoss further enforces the invariance of object features despite environmental changes, thereby improving the model's robustness in different scenes. These strategies collectively form a unified framework for ensuring contextual consistency within the same modality. Our method achieves state-of-the-art performance, surpassing previous approaches by +16.3 AP on OmniLabel and +14.9 AP on D^3. These results demonstrate the importance of enforcing intra-modal consistency, significantly enhancing model generalization in diverse environments. Our code is publicly available at: https://github.com/bozhao-li/CCL