Posted 2025-03-16Updated 2025-05-28Reviewa few seconds read (About 3 words)Scene Graph Generation by Iterative Message Passing
Posted 2025-03-14Updated 2025-05-28Reviewa few seconds read (About 86 words)RelTR= Relation Transformer for Scene Graph Generation RelTR是自下而上的方法, 使用基于Transformer的 object detector(例如DETR)生成对象候选者,然后使用relation transformer来预测object pairs之间的关系。它还设计了一种基于积分的关系表示方法,该方法将关系编码为二维矢量场。
Posted 2025-03-14Updated 2025-05-28Reviewa few seconds read (About 93 words)Fully Convolutional Scene Graph Generation One stage relation detection (detects objects and relations simultaneously). 这个模型受启发于 [[CenterNet]] 和 [[OpenPose Using Part Affinity Fields]],通过添加一个新的用于生成RAF的卷积头来获取物体之间的关系。 具体RAF是如何定义的,以及如何训练(Loss): Results
Posted 2025-03-14Updated 2025-05-28Reviewa few seconds read (About 3 words)Visual Relationship Detection with Language Priors
Posted 2025-03-13Updated 2025-05-28Reviewa few seconds read (About 0 words)Image generation from scene graphs
Posted 2025-03-13Updated 2025-05-28Reviewa few seconds read (About 3 words)Scene Graph Generation With Hierarchical Context
Posted 2025-03-13Updated 2025-05-28Reviewa few seconds read (About 20 words)Energy-Based Learning for Scene Graph Generation 这篇看不懂。。。 Can effectively utilize structural information in the output space
Posted 2025-03-13Updated 2025-05-28Reviewa few seconds read (About 0 words)Iterative Scene Graph Generation
Posted 2025-03-13Updated 2025-05-28Reviewa few seconds read (About 21 words)Image Retrieval using Scene GraphsBuilding an efficient structured representation that captures comprehensive semantic knowledge is a crucial step towards a deeper understanding of visual scenes
Posted 2025-03-13Updated 2025-05-28Reviewa few seconds read (About 6 words)From Pixels to Graphs= Open-Vocabulary Scene Graph Generation with Vision-Language Models