合成对象组合用于检测、分割和接地任务中的可扩展和准确学习
Synthetic Object Compositions for Scalable and Accurate Learning in Detection, Segmentation, and Grounding
AI总结：
合成对象组合（SOC）通过新的以对象为中心的组合策略，实现了准确且可扩展的数据合成流程，提升了检测、分割和接地任务的性能，尤其在低数据和封闭词汇场景中表现突出。
摘要：
视觉分组--通过实例分割、视觉接地和对象检测等任务进行操作--使应用范围从机器人感知到照片编辑。这些计算机视觉的基本问题由大规模、 painstakingly 注释的数据集推动。尽管其影响，这些数据集在构建成本高、覆盖范围有偏见且难以扩展。合成数据集提供了一个有前途的替代方案，但它们在灵活性、准确性和组合多样性方面存在困难。我们引入了合成对象组合（SOC），一种通过新的以对象为中心的组合策略来实现准确和可扩展的数据合成流程。它通过3D几何布局增强和相机配置增强，结合生成和谐化和掩码区域加权融合，将高质量的合成对象段组合成新的图像，从而产生准确且多样的掩码、框和提及表达。在仅使用10万合成图像训练的模型在性能上优于那些在更大真实数据集（GRIT 20M，V3Det 200K）和合成流程（Copy-Paste，X-Paste，SynGround，SegGen）上训练的模型，提升幅度为+24-36%--在LVIS上达到+10.9 AP，在gRefCOCO上达到+8.4 NAcc。除了通用的开放词汇设置外，SOC还能够为不同的使用场景构建可控的数据集，并在低数据和封闭词汇场景中提升性能。将LVIS和COCO与合成对象段增强，在不同真实数据规模下都能产生强大的性能，并在极有限的真实数据条件下产生更大的改进，包括在1%的COCO数据设置上达到+6.59 AP。此外，这种可控性使能够针对类内提及生成定向数据，这是一个我们提出的需求细粒度属性区分的诊断接地任务。
英文摘要：
Visual grouping -- operationalized through tasks such as instance segmentation, visual grounding, and object detection -- enables applications ranging from robotic perception to photo editing. These fundamental problems in computer vision are powered by large-scale, painstakingly annotated datasets. Despite their impact, these datasets are costly to build, biased in coverage, and difficult to scale. Synthetic datasets offer a promising alternative but struggle with flexibility, accuracy, and compositional diversity. We introduce Synthetic Object Compositions (SOC), an accurate and scalable data synthesis pipeline via a novel object-centric composition strategy. It composes high-quality synthetic object segments into new images using 3D geometric layout augmentation and camera configuration augmentation with generative harmonization and mask-area-weighted blending, yielding accurate and diverse masks, boxes, and referring expressions. Models trained on just 100K of our synthetic images outperform those trained on larger real datasets (GRIT 20M, V3Det 200K) and synthetic pipelines (Copy-Paste, X-Paste, SynGround, SegGen) by +24-36% -- achieving +10.9 AP on LVIS and +8.4 NAcc on gRefCOCO. Beyond the general open-vocabulary setup, SOC also enables controllable dataset construction for different use cases and boosts performance in both low-data and closed-vocabulary scenarios. Augmenting LVIS and COCO with synthetic object segments delivers strong performance across different real-data scales and yields even greater improvements under extremely limited real-data conditions, including +6.59 AP on a 1% COCO data setup. Furthermore, this controllability enables targeted data generation for intra-class referring, a diagnostic grounding task we propose that requires fine-grained attribute discrimination.
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。
