• willvarfar 10 months ago

    Completely newbie questions from someone outside the field who hasn't been following closely:

    Is the spatial memory a fixed size (how big?) or does it grow over time?

    And is there a point at which is is saturated and future results decline?

    • slashdave 10 months ago

      > Is the spatial memory a fixed size (how big?) or does it grow over time?

      Fixed number of parameters

      > And is there a point at which is is saturated and future results decline?

      The model is statistical, so precision will steady improve, asymptotically approaching a maximum

    • bahmboo 10 months ago

      From my understanding this is using the input images to render a new image from an arbitrary viewpoint. There may be a 3D reconstruction in the pipeline but this package produces a rendered image as the output. Anyone else see it this way? It’s very fast and cool for sure.

      • mab122 10 months ago

        Very interesting stuff. I wonder how this one camera (one viewpoint), flat images models work in completely novel environments (not seen in training data). I am wondering if this model could be used with stereo cameras as is.

        • moralestapia 10 months ago

          Two authors (so, one, lol)!

          Truly impressive, congrats to the young grad :D.

          • hahnchen 10 months ago

            > Two authors (so, one, lol)!

            What do you mean?

            • snovv_crash 10 months ago

              First one did the work, second one is the professor of the lab they are in.

          • RobotToaster 10 months ago

            From the previews I'm guessing this isn't going to be any use for 3d scanning?

            • ImHereToVote 10 months ago

              Can this result in a colmap dataset that can be used by Gaussian Splatting generation?

              • lelag 10 months ago

                There would not be much point. Colmap is already very capable in reconstructing a 3D scene from images from unknown poses if you have the camera intrinsics.

                Besides processing speed, this project (and the underlying dust3r model) strength is that it works with very few images. You basically just need 2, and it can infer pseudo instrinsics and matching extrinsics on it's own.

                I don't see why it could not be adapted to output gaussian splats instead. As a matter of fact, it's already been done with dust3r: https://github.com/nerlfield/wild-gaussian-splatting.

                • crubier 10 months ago

                  Colmap is also very slow for large scenes. Replacing Colmap with something faster would be a huge improvement for 3DGS pipelines. But Spann3r isn't there yet imo

                • vessenes 10 months ago

                  This relies on Dust3r underneath as part of its stack (I didn’t read carefully enough to tell you if it’s training or inference but I think it’s training), which outputs splats. What’s special about this is that it outputs really dense nice point clouds with arbitrary photos. We have a lot more tools that work well with point clouds than with splats, so this is nice work.

                • mitthrowaway2 10 months ago

                  Interesting! How well does it handle scenes containing moving objects?

                  • esther598 10 months ago

                    [dead]