At Open Sauce, main hall.

The Fujifilm booth has a gaussian splatting demo with automatic collisions. The person at the booth also explained to me how the pop in artifacts related to reflections have been recently fixed in previous gaussian splatting research over the last few months and that it has come a long way.

I asked if their gaussian splatting demo was limited to static scenes, so they also showed a dynamic demo, where the gaussian splatting was being used for a person (themselves, actually) talking and moving around.

They also pointed out to us that it is surprising that the gaussian splatted scene output is often much smaller than the input files, but as a group we didnt find this too surprising since we imagined there being duplicate or common information between all the input scenes.