用无标注图像提升小样本目标检测，不依赖外部标签库。
SimLTD: Simple Supervised and Semi-Supervised Long-Tailed Object Detection
- 先用大量头部类别预训练，再迁移学习尾部小样本类。
- 在LVIS v1上，监督与半监督均创最佳性能纪录。
- 无需人工标注，适合真实场景中小样本检测任务。
尽管现代视觉识别系统取得显著进展，但仍难以从少量样本中学习。本文聚焦于物体类别呈长尾分布下的目标检测任务。现有方法依赖ImageNet等大型标注数据集来增强少样本实例，但在实际应用中受限。我们提出一种灵活可扩展的方法，利用容易获取的无标注图像，无需额外人工标注。SimLTD框架简单直观，包含三步：(1) 在丰富头部类别上预训练；(2) 在稀缺尾部类别上进行迁移学习；(3) 在采样的头尾类别组合上微调。该方法无需元学习或知识蒸馏等复杂机制，仅通过引入无标注图像，在监督与半监督设置下于挑战性LVIS v1基准上均取得新纪录性能。
原文摘要 · Abstract (English)
While modern visual recognition systems have made significant advancements, many continue to struggle with the open problem of learning from few exemplars. This paper focuses on the task of object detection in the setting where object classes follow a natural long-tailed distribution. Existing methods for long-tailed detection resort to external ImageNet labels to augment the low-shot training instances. However, such dependency on a large labeled database has limited utility in practical scenarios. We propose a versatile and scalable approach to leverage optional unlabeled images, which are easy to collect without the burden of human annotations. Our SimLTD framework is straightforward and intuitive, and consists of three simple steps: (1) pre-training on abundant head classes; (2) transfer learning on scarce tail classes; and (3) fine-tuning on a sampled set of both head and tail classes. Our approach can be viewed as an improved head-to-tail model transfer paradigm without the added complexities of meta-learning or knowledge distillation, as was required in past research. By harnessing supplementary unlabeled images, without extra image labels, SimLTD establishes new record results on the challenging LVIS v1 benchmark across both supervised and semi-supervised settings.
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。