Clean pdf to html conversion

Question

I am searching for a clean conversion from pdf to html or xml. I know that there are many solutions, which follow the way of keeping the positions and layouts of several contents. But I am searching for a conversion tool, which converts into clean html without css or into xml. Existing products create a mess of thousand div-tags and span-tags, but you cannot differentiate between a header and a table.
Example (what I need):
<h1>This is a header 1</h1>
<p>This is text</p>
<img src="..... />
<h2> This is header 2 </h2>
<table>.....This is a table ....... </table>
Existing solutions:
<div style="position:.....">Text</div>
<div style="position:.....">Text</div>
<div style="position:.....">Text</div>
<div style="position:.....">Text</div>
Is there any product, which can do that? (batch conversion on servers (e.g. JAVA))

Answer 1

Given that CSS is part of HTML - I don't see why that would be an issue.
Since this is Adobe's forum, we offer a Java-focused server side solution called LiveCycle ES.