\

Lift4D: Harmonizing Single-View 3D Estimation for 4D Reconstruction In-the-Wild

97 points - today at 2:40 PM

Source
  • b1n

    today at 9:28 PM

    Enemy of the State predicted this: https://www.youtube.com/watch?v=3EwZQddc3kY&t=11s

    It's only a matter of time before it's in use by law enforcement: https://news.ycombinator.com/item?id=48520807

    • bensmoif

      today at 5:38 PM

      Really interesting, and a link to your GitHub with a "code coming soon". Please get us the tool already!

      • christianh

        today at 6:29 PM

        Looks cool. How’s it different from https://github.com/gaomingqi/sam-body4d ? My guess: Is sam-body4d focused on temporally consistent body reconstruction into an MHR model, and Lift4D creates the full scene and a model of the tracked object (that doesn’t need to be a human). I may be way off, but a comparison would help me!

        • fraywing

          today at 9:03 PM

          wondering how accurate the extrapolated distances are? Like the Rhino and the tree -- I wonder if the approximation could be useful in something like forensics from a security video?

          • darepublic

            today at 3:46 PM

            I'm sorry but this just reminds me of this tng episode that inspired me as a teen https://www.youtube.com/watch?v=4Faiu360W7Q

          • poly2it

            today at 4:25 PM

            This is insanely promising. What a time to be alive!

              • Joel_Mckay

                today at 4:39 PM

                Two Minute papers fan? fun YT channel =3

            • tamimio

              today at 4:45 PM

              Cool, so now when the swarm drones come after you, they can reconstruct the video on the fly from that single CCTV shot taken by flock, and pinpoint your location from the aggregated data by plantir!