近年来，许多国外的儿童绘本被引进和翻译到中国的图书出版市场。但是儿童绘本的翻译热情并不意味着这些绘本都能得到高质量的翻译，这也是国内儿童绘本商业化的操作模式导致的。因此针对儿童绘本的翻译的研究对解决这个问题至关重要，然而该领域的研究仍处在初步发展阶段，翻译策略的系统研究还没有开展起来。由于儿童绘本具有多模态的特点，本研究试图从多模态的角度探讨儿童绘本的翻译策略，以期为儿童绘本的翻译实践提供初步参考。通过研究《羊驼拉玛》系列双语绘本，运用多模态的视角来考察儿童绘本的英汉翻译策略，分析语言文字和视觉图像之间的复杂关系，并借鉴最新的文本分析模型“新跨媒介关系框架模型”，并采用定性分析的方法，构建了一个儿童绘本翻译的文本分析模型。从分析中可以看出，翻译绘本时主要运用了三种有效的翻译策略：增译、缩译和异化。当图像与文字文本处于互补关系时，翻译时应采用增译策略以补充完整多模态文本的信息，部分文字文本外的额外信息由图像提供；当图像与文字文本处于平等关系时，当两种模态的信息重复导致了信息冗余，翻译时应采用缩译策略以保持语言的简洁；当源文本呈现的一些特定文化现象或事物在译语文化中不存在时，异化策略成为了优先选择。

In recent years, many foreign children’s picturebooks have been introduced and translated into China’s book publishing market. Yet it is noteworthy that enthusiasm for translation fail to guarantee the high-quality of picture books, which can blame to the commercialization of the market. Research into the translation of children’s picturebooks is therefore essential to address this issue, but the related research is still at an early stage of development and systematic study into translation strategies has yet to be undertaken. As children’s picturebooks are multimodal in nature, this thesis attempts to explore the translation strategies of children’s picturebooks from a multimodal perspective, in the hope of providing preliminary reference for the translation practice of children’s picturebooks. By doing a case study of Llama Llama series, this thesis applies a multimodal perspective to examine the English-Chinese translation strategies of children’s picturebooks. By analysing the complex relationship between verbal text and visual images and drawing on the latest model of text analysis “New COSMOROE Model”, this thesis adopts qualitative analysis to construct a textual analysis model for the translation of children’s picturebooks. From the analysis, it is clear that there are three valid picturebook translation strategies: addition, condensation and foreignization. When the images are in a complementary relationship with the text, the addition strategy should be used to complete the information of the multimodal text as the visual images manage to provide information that is not included in the text; when the images are in an equal relationship with the text, and the repetition of the information in both two modalities results in redundancy, condensation strategy should be used to preserve language simplicity; when the source text presents some cultural phenomenon or culture-specific items that do not exist in the target language culture, foreignization strategy becomes the prior choice.