While modern adaptations of Chinese classics have drawn keen scholarly interests lately, the comic adaptation of Chinese traditional poetry remains under-investigated. Extending the previous research on intersemiotic translation and comics, this paper, drawing on the analytical framework of systemic functional semiotics, examines distribution of process types of language in poems in comparison with that in comic images in the exemplary case drawn by Cai Zhizhong, using UAM image as the annotation tool. The comic book formulates a multimodal corpus that consists of 1,097 clauses and 605 images. We have manually analyzed the process type of the poems and their corresponding comic panels. Our quantitative and qualitative results show that there are distinct patterns of process-type distributions between verbal poems and images. Poems have been turned into perceptions, actions, and verbal processes in comic strips, which serve various purposes such as construction of the poet's gaze, relations building, storyline development, dramatization, metaphor visualization, etc. The paper is concluded with discussion on how the intersemiotic translation of poems might produce effects on readers.