Extracting only the background

[TomLucidor]

If someone wants to extract the background context for analysis, what can be done? This is a bit different from unCLIP and that text can be extracted rather than being a CLIP-based context https://github.com/pythongosssss/ComfyUI-WD14-Tagger

Side note: this also applies to RemBg for real life photos https://github.com/Jcd1230/rembg-comfyui-node