Abstract: Infant emotions significantly influence their cognitive development and self-awareness, playing a crucial role in their growth. The field of infant emotion research currently confronts challenges such as insufficient emotion datasets, non-uniform quantification standards, and limited accuracy in emotion recognition. Therefore, effectively recognizing infant emotions has become a central issue in this field. This paper reviews and summarizes current research on infant emotion recognition, focusing on infant emotion datasets, representation models, and classification methods across various modalities. Initially, it enumerates and introduces available public datasets. Subsequently, it analyzes common models for representing infant emotions and methods for classifying them. Finally, it outlines existing challenges and discusses future research directions, aiming to guide further studies.

Key words: infant emotions, emotion datasets, infant emotion representation models, infant emotion recognition

摘要： 婴儿情感在塑造婴儿的认知和自我意识中扮演着关键角色，对婴儿成长具有至关重要的影响。目前婴儿情感研究领域，面临婴儿情感数据集不足、情感量化标准不统一，以及情感识别的准确性受限等问题。因此，如何有效地识别婴儿情感已成为该领域当前研究的核心难题。通过梳理当前婴儿情感识别领域的研究现状，从婴儿情感数据集、婴儿情感表示模型以及婴儿情感识别方法等方面对婴儿情感识别研究进行了分析与总结。列举并介绍了当前可用的公开数据集。总结并分析了常见婴儿情感模型，以及不同模态下的婴儿情感识别方法。概括该领域当前存在的问题以及探讨未来的研究方向，旨在为进一步的研究提供方向。

关键词: 婴儿情感, 情感数据集, 婴儿情感表示模型, 婴儿情感识别