基于新型可供性类型的零样本二维定位
Zero-shot 2D Grounding with Novel Affordance Types
- Purdue University（普渡大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
本文提出新型可供性类型的零样本二维定位任务，构建对应基准，提出AffordAnything及可训练变体AffordAnything+，在AGD20K-NAT基准上较SOTA方法OOAL的IoU@0.4提升12.3%。
AI中文摘要：
二维可供性定位旨在定位人类可与物体交互的区域。现有研究聚焦于识别训练期间所见的可供性类型，未探究模型对新型可供性的泛化能力，而这对实际应用至关重要。我们提出了新型可供性类型（NAT）的零样本二维定位任务，并引入了NAT基准。随后，我们提出了AffordAnything这一无训练方法，其利用分割线索，动机源于可供性区域与物体子部件间的强相关性。为进一步提升性能，我们开发了可训练变体AffordAnything+，该变体学习结合这些线索。在我们提出的AGD20K-NAT基准上，我们的最佳模型AffordAnything+相较于SOTA可供性定位方法OOAL，在IoU@0.4指标上实现了12.3%（绝对）的显著提升。
英文摘要：
2D affordance grounding aims to locate the region of an object that a human can interact with. Existing research focuses on recognizing affordance types seen during training and does not study models' ability to generalize to novel affordances, which is crucial for real-world applications. We propose the task of zero-shot 2D grounding with novel affordance types (NAT) and introduce the NAT benchmarks. We then propose AffordAnything, a training-free method that leverages segmentation cues, motivated by the strong correlation between affordance regions and object subparts. To further improve performance, we develop AffordAnything+, a trainable variant that learns to combine these cues. On the proposed AGD20K-NAT benchmark, our best model AffordAnything+ achieves a substantial improvement of 12.3% (absolute) in IoU@0.4 over the SOTA affordance grounding method, OOAL.
