I made a post last year wondering something similar—why despite the current boon of film there isn't much beyond flatbeds and the Plustek/PI/Braun scanners that nobody seems to really love. I think what was surmised was that despite the film resurgence, it's a miniscule fraction of the market (and near nonexistent in commercial use beyond, say, minilabs) so there's no real impetus to invest in building new ones, and there's plenty old ones out there that still do work. Just going on other film photographers I know, they either shell out for scans with processing at labs, or DSLR scan.
I, for one, would love a new dedicated film scanner like the old Coolscans, with an automatic transport and multiple formats. (I'd also love a new MF camera that isn't a 3D-printed box camera, but the only option, the Rollei Hy6 II, is just a bit out of my price range!) But I don't really foresee any of the big manufacturers going for it at the moment. I would imagine for Epson and Canon, the flatbed market is small enough as it is, and they're happy building something double-duty for those of us that still scan at home.
I've been DSLR scanning for a little over a year now, with far better results than I got with my Coolscan, but it's been a massive time investment—I built a few prototype negative holders and light sources before waiting many months for the Negative Supply Kickstarter, which to date I think is the best setup for volume scanning. Then I ended up going through several different light pads/tables trying to find an adequate source, and finally ditching the tripod for a copy stand of my own design. Add to that the back-end kludge of running a script to flip the levels in C1 and trying to remember which adjustments work in reverse or not.
Phenomenal results in the end, and I have some sense of pride building my setup, but it would be so much nicer to have a single box to stick negatives into and spits out DNG positives. It's like rolling your own developer—some like the tinkering, but others don't want to bother, and that's okay.
I could see someone enterprising building a fully automated machine that one slots a user-supplied body like a Fuji into, with an accompanying control software suite and offering that on Kickstarter or something. Plenty of enterprising people have made little apps to do various component functions like automated capture, DNG conversion, etc, it would just take putting it all together.
But that, I would guess, would be massively expensive. See the Filmomat—I tinkered a bit with Arduino kits (before losing interest!) and my first thought was building an automated film processor. Someone did that, but it's not even close to cost effective, and doesn't have the wide range of accessories as, say, the old Jobo Autolabs.
Not being pessimistic, just thinking out loud. I do think DSLR scanning will be what takes off, with a cottage industry around accessories like NS is doing. The big advantage is most photogs already own the most expensive component, the sensor.
As a coda, in relation to new cameras—I suppose the irony of film photography these days is we have so much technology to build with, like 3D printing/CNC to build out a large format camera from scratch, and digital editing and printing that rivals darkroom prints, but without the heft of a giant multi-national company, we won't see something as electro-mechanically advanced as the F6.
