Repository navigation
Significant quality loss in NuRec (OmniNuRecFieldAsset) rendering vs SuperSplat for Lyra-generated dynamic 3DGS scenes #680
|
6.0.0 Operating System Ubuntu 24.04 Detailed Description I’m reconstructing a dynamic scene from a single fixed-position 1080p/30fps video using NVIDIA’s ViPE → GEN3C → Lyra pipeline. Lyra outputs standard 3DGS .ply files at several timestamps throughout the video (“bullet time” snapshots), each containing ~2M Gaussians. What works: When I load the raw gaussians_0.ply output directly in SuperSplat, the scene renders clearly. What doesn’t work: I converted the same fixed .ply to NuRec USDZ using: python -m threedgrut.export.scripts.ply_to_usd <input.ply> --output_file <output.usdz> When I load the resulting .usdz in Isaac Sim, the OmniNuRecFieldAsset prim (with density_field and emissive_color_field) is created correctly in the stage, but the rendered result is significantly blurrier/streakier than the SuperSplat view of the identical (opacity-fixed) data even at the same approximate camera angle, and even after letting RTX Real-Time 2.0 accumulate frames. I also tried the alternative ParticleField3DGaussianSplat (UsdVol “standard”/“lightfield”) schema by manually authoring the prim myself (positions, scales as exp(scale_i), orientations as normalized quaternions, opacities as sigmoid(opacity), SH degree-0 coefficients from f_dc_0/1/2, with required extent attribute). That path rendered (after fixing missing extent/Boundable requirements), but more blurry than NuRec, with worse streak/noise artifacts when rotating the camera further from the source viewpoint, consistent with what’s described in the other thread comparing these two pipelines (forum thread 373260). So for now I am sticking with NeRec Questions: Is there a known minimum angular/baseline coverage requirement for Lyra-generated 3DGS scenes to reconstruct cleanly across a wide viewing range, or is this purely a function of how many/how-wide the GEN3C trajectories are? |
Replies: 2 comments 3 replies
|
Could you Please attach a minimal PLY/USDZ sample and the exact export commands for us to replicate your issue? I will also reach out to the internal teams about your questions. |
|
Thanks @Aryav-T I will create an internal ticket for the quality loss issue you mentioned. Will you be able to share your PLY example with us? As for getting meshes from the lyra-1 output, I am still following up with the internal team for potential options. |


@Aryav-T About getting mesh from lyra-1 output:
For meshing, current research suggests that a conversion of a 3DGS asset to a mesh "after the fact" will lead to a lower quality compared to more integrated approaches that produce a mesh directly.
Since Lyra only produces 3DGS, you may try some multi-step approach which is not perfect but may prove to be satisfactory for your use case.
One approach is going 3DGS -> rendered depth maps -> fusion into a truncated signed distance function (TSDF) -> marching cubes.
It seems the public 2DGS codebase provides a similar approach, though likely the Lyra asset may not be directly compatible.