arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1908.06336cs.CVcs.AIcs.CLcs.LG
收藏

VQA中的简单空间语言功能需要什么?

What is needed for simple spatial language capabilities in VQA?

University of Cambridge(剑桥大学)

Alexander Kuhnle, Ann Copestake

摘要:

视觉问答(VQA)包括多种语言能力。诊断基准数据集CLEVR通过帮助更好地评估和区分体外计数、比较和空间推理等基本能力的模型,推动了研究进展。按照这种方法,我们专注于空间语言能力和调查的问题:什么是处理简单的视觉空间关系的关键成分?我们研究了SAN、RelNet、FiLM和MC模型,并评估了它们在诊断数据上的学习行为,这些诊断数据只关注空间关系。通过比较分析和有针对性的模型修改,我们确定了在CNN-LSTM基线基础上进行实质性改进所真正需要的内容。

英文摘要:

Visual question answering (VQA) comprises a variety of language capabilities. The diagnostic benchmark dataset CLEVR has fueled progress by helping to better assess and distinguish models in basic abilities like counting, comparing and spatial reasoning in vitro. Following this approach, we focus on spatial language capabilities and investigate the question: what are the key ingredients to handle simple visual-spatial relations? We look at the SAN, RelNet, FiLM and MC models and evaluate their learning behavior on diagnostic data which is solely focused on spatial relations. Via comparative analysis and targeted model modification we identify what really is required to substantially improve upon the CNN-LSTM baseline.

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。

↑