This is an improved method. The original post is here
Based on the original working principle, the workflow has been optimized. Originally, two LoRAs (ColorManga and Anime2Realism) were needed, but now only Anime2Realism is required. The prompt words and various parameters in the current workflow are the optimal solutions after extensive testing, so it is not recommended to modify them. To use this workflow, you just need to upload a 3D image -- then run it -- and wait for the result. It's that simple. The following is a demonstration of the effect images.
the LoRA link:
https://civitai.com/models/1934100
the workflow link
Convert 3D images into realistic photos | Patreon
