VR tools like Mantra, Cara, and Mocha are are basically video manipulation but in VR, their 2D counterparts might be really simple and accessible but getting the same effect in VR is a completely different level of difficulty. There are a lot of things that can make a manipulated 2d image look real, simple photoshopping can fool the eyes really well. As long as the scale and lighting looks correct, our eyes would be happy with the image, but in VR everything have to be correct for both eyes, therefore way more information is needed and way more work is needed to make things look correct for both eyes. Tricks that can fool the eyes in traditional photo manipulation can no longer work when they are perceived with depth. 
I think this GIF image is a great example of how something might look perfect when only viewed at one perspective that is determined by the photographer, while it would have not looked the same if the same effect is viewed with depth or at another perspective. This is why all three companies prioritizes the accuracy of their effects when they migrate these traditional tools into a new VR environment.
The state of VR today is split into film and CGI, and for the best visual effects we try to merge them just like the jack black Goosebumps example where he was filmed with a 360 camera but later perfected with CGI and 3d models of monsters. The two different fields of VR technology is actually blended for the best effect, to create something that is accurate and correct for the human eye.