先定位目标再识别，提升模型在复杂背景下的鲁棒性
Bringing the Context Back into Object Recognition, Robustly
- 用零样本检测先定位前景，再进行识别
- 在多种数据集上显著提升识别准确率，尤其对长尾背景更稳健
- 适合需要高鲁棒性的实际部署场景，如自动驾驶
在目标识别中，目标本身（前景，FG）和其周围环境（背景，BG）均可能起重要作用。然而，标准监督学习常导致模型过度依赖背景，限制了在真实场景中的鲁棒性。现有方法多通过抑制背景来改善泛化，但牺牲了上下文信息。本文提出“局部定位以稳健识别”（L2R2），利用零样本检测技术在识别前定位前景，既保持上下文感知能力，又增强对分布偏移的鲁棒性。该方法在标准监督识别和多模态零样本识别（使用VLMs）中均有提升，且对长尾背景和分布变化具有更强适应性。实验表明，在多种数据集上定位先行是可行的，但在部分数据集上仍受限于目标检测性能。
原文摘要 · Abstract (English)
In object recognition, both the subject of interest (referred to as foreground, FG, for simplicity) and its surrounding context (background, BG) may play an important role. However, standard supervised learning often leads to unintended over-reliance on the BG, limiting model robustness in real-world deployment settings. The problem is mainly addressed by suppressing the BG, sacrificing context information for improved generalization. We propose "Localize to Recognize Robustly" (L2R2), a novel recognition approach which exploits the benefits of context-aware classification while maintaining robustness to distribution shifts. L2R2 leverages advances in zero-shot detection to localize the FG before recognition. It improves the performance of both standard recognition with supervised training, as well as multimodal zero-shot recognition with VLMs, while being robust to long-tail BGs and distribution shifts. The results confirm localization before recognition is possible for a wide range of datasets and they highlight the limits of object detection on others
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。