3 papers
cs.GR2026
Directed Distance Fields for Constant-Time Ray Queries on Gaussian Splatting
Subhankar MIshra
3D Gaussian Splatting (3DGS) renders new views of a scene in real time. Like every rasterizer, it answers only primary rays, the rays from the camera through the image. It cannot t…
cs.LG2026
mHC-GNN: Manifold-Constrained Hyper-Connections for Graph Neural Networks
Subhankar Mishra
Graph Neural Networks (GNNs) suffer from over-smoothing in deep architectures and expressiveness bounded by the 1-Weisfeiler-Leman (1-WL) test. We adapt Manifold-Constrained Hyper-…
cs.CV2026
CleanGS: Pareto-Guided Post-Hoc Cleanup of 3D Gaussian Splatting Scenes
Subhankar Mishra, Jyotirmaya Shivottam, Om Kumar
A pretrained 3D Gaussian Splatting (3DGS) reconstruction can reproduce its capture views well while retaining off-object structure that becomes visible under reframing. We present…