Human-Object Interaction (HOI) detection is an important problem to
understand how humans interact with objects. In this paper, we explore
interactiveness knowledge which indicates whether a human and an object
interact with each other or not. We found that interactiveness knowledge can be
learned across HOI datasets and bridge the gap between diverse HOI category
settings. Our core idea is to exploit an interactiveness network to learn the
general interactiveness knowledge from multiple HOI datasets and perform
Non-Interaction Suppression (NIS) before HOI classification in inference. On
account of the generalization ability of interactiveness, interactiveness
network is a transferable knowledge learner and can be cooperated with any HOI
detection models to achieve desirable results. We utilize the human instance
and body part features together to learn the interactiveness in hierarchical
paradigm, i.e., instance-level and body part-level interactivenesses.
Thereafter, a consistency task is proposed to guide the learning and extract
deeper interactive visual clues. We extensively evaluate the proposed method on
HICO-DET, V-COCO, and a newly constructed PaStaNet-HOI dataset. With the
learned interactiveness, our method outperforms state-of-the-art HOI detection
methods, verifying its efficacy and flexibility. Code is available at
this https URL

本文研究了人物和物体间的交互行为，发现交互知识可以跨越不同的数据集，形成通用的交互知识网络，并且可以与任何 HOI 检测模型合作，进而提高 HOI 检测效率和准确性。通过使用实例和人体部位的特征信息，本文还提出了一种层次化的交互知识获取方法，同时采用一致性任务深度提取可视化线索。在 HICO-DET、V-COCO 和 PaStaNet-HOI 数据集上的实验结果表明，该方法优于现有的 HOI 检测方法。

人物 - 物体交互检测的可传递交互知识

Transferable Interactiveness Knowledge for Human-Object Interaction  Detection

We present HARP, a novel method for learning low dimensional embeddings of a
graph's nodes which preserves higher-order structural features. Our proposed
method achieves this by compressing the input graph prior to embedding it,
effectively avoiding troublesome embedding configurations (i.e. local minima)
which can pose problems to non-convex optimization. HARP works by finding a
smaller graph which approximates the global structure of its input. This
simplified graph is used to learn a set of initial representations, which serve
as good initializations for learning representations in the original, detailed
graph. We inductively extend this idea, by decomposing a graph in a series of
levels, and then embed the hierarchy of graphs from the coarsest one to the
original graph. HARP is a general meta-strategy to improve all of the
state-of-the-art neural algorithms for embedding graphs, including DeepWalk,
LINE, and Node2vec. Indeed, we demonstrate that applying HARP's hierarchical
paradigm yields improved implementations for all three of these methods, as
evaluated on both classification tasks on real-world graphs such as DBLP,
BlogCatalog, CiteSeer, and Arxiv, where we achieve a performance gain over the
original implementations by up to 14% Macro F1.

提出了用于学习保留更高阶结构特征的图形节点低维嵌入的 HARP 方法，通过在嵌入之前压缩输入图来实现这一点，并使用简化的图形来学习一系列的初始表示，这些方法可以扩展到所有现有的图形嵌入神经算法，包括 DeepWalk, LINE 和 Node2vec。