Intentonomy: a Dataset and Study towards Human Intent Understanding

Abstract

An image is worth a thousand words, conveying information that goes beyond the physical visual content therein. In this paper, we study the intent behind social media images with an aim to analyze how visual information can help the recognition of human intent. Towards this goal, we introduce an intent dataset, Intentonomy, comprising 14K images covering a wide range of everyday scenes. These images are manually annotated with 28 intent categories that are derived from a social psychology taxonomy. We then systematically study whether, and to what extent, commonly used visual information, i.e., object and context, contribute to human motive understanding. Based on our findings, we conduct further study to quantify the effect of attending to object and context classes as well as textual information in the form of hashtags when training an intent classifier. Our results quantitatively and qualitatively shed light on how visual and textual information can produce observable effects when predicting intent.

arXiv commentCVPR2021

References in corpus (9)

• Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings
• Word Embeddings Quantify 100 Years of Gender and Ethnic Stereotypes
• Deep Learning is Robust to Massive Label Noise
• Approximating CNNs with Bag-of-local-Features models works surprisingly well on ImageNet
• Attenuating Bias in Word Vectors
• Equal But Not The Same: Understanding the Implicit Relationship Between Persuasive Images and Text
• Recognition in Terra Incognita
• MIDAS: A Dialog Act Annotation Scheme for Open Domain Human Machine Spoken Conversations
• Predicting the Politics of an Image Using Webly Supervised Data