Hi,
Yes, you can definitely do this without using a server. It'll probably be easier using the Document APIs instead of PDFNet APIs, and faster too because you don't need to load the "full API".
Note that from a browser canvas object you can use the toDataURL or toBlob functions to get the image data
That doc object is the exact same as what you get from docViewer.getDocument() so then you can call doc.loadCanvasAsync to get the rendered page.
Let me know if you have any other questions about this.
Matt Parizeau
Software Developer
PDFTron Systems Inc.