Not sure if I requested the feature in a correct place. I have a case where I wanted to convert doc to html (text only). Right now my own solution is replacing all IMG tag in html and delete the file manually. It would be better if there is an option in unoconv to skip images as there will be less overhead (no need to save the image to disk then delete it instantly).
If the HTML export filter does not allow you to do this, unoconv wouldn't either. At least not out of the box.
There's a possibility that we allow users to write small scriptlets to perform changes to a document before converting, but that functionality does not exist yet. On the other hand, since unoconv is a python script anyhow, you can already modify the body of the script and add specific python code to manipulate the document before converting. (i.e. remove all images, merge changes, replace content, add index, etc...) But you need some python skills and a lot of perseverance to understand UNO :-/