arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.16083cs.CV
收藏

全卷积场景图生成

Fully Convolutional Scene Graph Generation

Futurewei Technologies Inc.(未来网络科技公司)

Hengyue Liu, Ning Yan, Masood S. Mortazavi, Bir Bhanu

摘要:

本文提出了一个完全卷积场景图生成(FCSGG)模型,同时检测对象和关系。大多数场景图生成框架使用预训练的两阶段对象检测器,如Faster R-CNN,并使用限界框特征构建场景图。这种流水线通常具有大量的参数和低推理速度。与这些方法不同,FCSGG是一种概念上优雅且高效的自底向上方法,其将对象编码为限界框中心点,并将关系编码为称为关系亲和域(RAF)的2D向量场。RAFs编码语义和空间特征,并明确表示一对对象之间的关系的子区域上的积分,从主题指向对象。FCSGG仅利用视觉特征,仍然为场景图生成强大的结果。在Visual Genome数据集上的综合实验证明了该方法的有效性、高效性和通用性。FCSGG在召回率和zero-shot召回率方面取得了非常有竞争力的结果,并显着减少了推理时间。

英文摘要:

This paper presents a fully convolutional scene graph generation (FCSGG) model that detects objects and relations simultaneously. Most of the scene graph generation frameworks use a pre-trained two-stage object detector, like Faster R-CNN, and build scene graphs using bounding box features. Such pipeline usually has a large number of parameters and low inference speed. Unlike these approaches, FCSGG is a conceptually elegant and efficient bottom-up approach that encodes objects as bounding box center points, and relationships as 2D vector fields which are named as Relation Affinity Fields (RAFs). RAFs encode both semantic and spatial features, and explicitly represent the relationship between a pair of objects by the integral on a sub-region that points from subject to object. FCSGG only utilizes visual features and still generates strong results for scene graph generation. Comprehensive experiments on the Visual Genome dataset demonstrate the efficacy, efficiency, and generalizability of the proposed method. FCSGG achieves highly competitive results on recall and zero-shot recall with significantly reduced inference time.

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。

↑