DL3DV
A large-scale 3D reconstruction dataset containing numerous indoor and outdoor scenes, used to evaluate neural rendering and 3D reconstruction methods.
Timeline (last 90 days)
On the DL3DV and RealEstate10K datasets, PPMem outperforms previous 3D-grounded world models in video quality while jointly outputting metric-scale depth, surface normals, and exportable plane geometry.
DL3DV provides multi-view videos of large-scale real indoor and outdoor scenes with wide viewpoint spans and complex lighting; RealEstate10K comes from real estate videos and is characterized by continuous camera trajectories through indoor corridors and rooms.
The MFGS framework embeds multi-view feature encoding directly into the feed-forward 3DGS pipeline, achieving up to 99.86% bitrate savings (BD-rate) on the three datasets RE10K, ACID and DL3DV, with about a 12x end-to-end speedup.
All Facts (3)
On the DL3DV and RealEstate10K datasets, PPMem outperforms previous 3D-grounded world models in video quality while jointly outputting metric-scale depth, surface normals, and exportable plane geometry.
50%UnverifiedDL3DV provides multi-view videos of large-scale real indoor and outdoor scenes with wide viewpoint spans and complex lighting; RealEstate10K comes from real estate videos and is characterized by continuous camera trajectories through indoor corridors and rooms.
50%UnverifiedThe MFGS framework embeds multi-view feature encoding directly into the feed-forward 3DGS pipeline, achieving up to 99.86% bitrate savings (BD-rate) on the three datasets RE10K, ACID and DL3DV, with about a 12x end-to-end speedup.
50%