两篇新研究论文CAGE-SGG和ReLIC-SGG提出了用于开放词汇场景图生成的新颖方法。CAGE-SGG侧重于使用反事实证据验证预测的关系，以确保它们在视觉上是合理的，而不是依赖于语言先验。ReLIC-SGG通过将未标注的关系视为潜在变量并构建语义关系格来推断缺失的连接，从而解决了标注不完整的问题。 AI
影响 引入了更可靠和可解释的视觉场景理解的新技术，可能改进下游AI应用。
排序理由 该集群包含arXiv上的两篇关于场景图生成新方法的学术论文。
AI 生成摘要 · Google Gemini · 来自 4 个来源。 我们如何撰写摘要 →
arXiv:2604.22274v1 Announce Type: new Abstract: Open-vocabulary scene graph generation (SGG) aims to describe visual scenes with flexible and fine-grained relation phrases beyond a fixed predicate vocabulary. While recent vision-language models greatly expand the semantic coverag…
arXiv:2604.22546v1 Announce Type: new Abstract: Open-vocabulary scene graph generation (SGG) aims to describe visual scenes with flexible relation phrases beyond a fixed predicate set. Existing methods usually treat annotated triplets as positives and all unannotated object-pair …
Open-vocabulary scene graph generation (SGG) aims to describe visual scenes with flexible relation phrases beyond a fixed predicate set. Existing methods usually treat annotated triplets as positives and all unannotated object-pair relations as negatives. However, scene graph ann…
Open-vocabulary scene graph generation (SGG) aims to describe visual scenes with flexible and fine-grained relation phrases beyond a fixed predicate vocabulary. While recent vision-language models greatly expand the semantic coverage of SGG, they also introduce a critical reliabi…