Which Acrobat SDK Technology should I use?

I thought that plug-in technology would be the slam-dunk answer to my application needs, but was disappointed when I realized that the x64 third party library that I want to use with my plug-in will not work with Win32.  So how can I get it done?

The following is an incomplete list of what I want my application to do:

1. Give the user a UI/window in which to interact with PDF pages.  This would preferably be Acrobat.
2. Provide the user with drawing tools with which the user can draw lines, boxes, circles, etc. on pages and have those saved in a layer, with zoom capabilities, for that page.
3. The user will have access to a "snipping tool", similar to Microsoft's Snipping Tool, used to capture and save images from PDF pages to a library/list of images.
4. The user will be able to search for text and images on other yet unopened PDF document pages.
5. The user will be able to use the image library to search for images on pages; including images within images that are on pages.  Here, I assume that the PDF document pages will be converted to binary image format and stored in memory.  The search would be in memory rather than on disk.
6. Machine learning would be used to search for "similar" text and images within documents.  This is where the above mentioned x64 third party library would come into play.
7. The user would maintain a list of document filenames which are of interest.
8. Results of successful searches would record data such as document names, text/image coordinates and other meta data to a database.
9. The app must utilize multicore processing and GPU acceleration when available.

I do not believe that a plug-in alone would give me what I need, especially considering that the third party library is compiled as x64.  IAC looks interesting, but, again, doesn't allow all that I need. 

Would some combination of plug-in + IAC + PDF Library be an answer?  Or am I overlooking something that would make this task a lot easier?