OP2GS adds instance identities and dual opacities to 3D Gaussians so that visual rendering and object-mask rendering are handled by separate opacity channels, reducing label contamination while attaching semantics at the object level.
Berg, Wan-Yen Lo, Piotr Dollar, and Ross Girshick
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
citation-role summary
background 2
citation-polarity summary
fields
cs.CV 5verdicts
UNVERDICTED 5roles
background 2polarities
background 2representative citing papers
GuardMarkGS unifies watermarking and adversarial edit deterrence into a single optimization framework for protecting 3D Gaussian Splatting assets.
CAFE benchmark reveals that promptable segmentation models often produce correct masks for misleading prompts, showing a gap between localization accuracy and true concept understanding.
Hide-and-Seek Attribution combined with a diffusion autoencoder converts coarse vertebra-level labels into accurate lytic and blastic lesion segmentations, reaching F1 scores of 0.91 and 0.85 without any mask supervision.
Metric depth is recovered by expressing scale variations as a linear combination of basis maps from MDE cues whose weights are fit by least-squares to sparse metric anchors.
citing papers explorer
-
From Pixels to Concepts: Do Segmentation Models Understand What They Segment?
CAFE benchmark reveals that promptable segmentation models often produce correct masks for misleading prompts, showing a gap between localization accuracy and true concept understanding.
-
Learning Image-Adaptive Scale Fields for Metric Depth Recovery
Metric depth is recovered by expressing scale variations as a linear combination of basis maps from MDE cues whose weights are fit by least-squares to sparse metric anchors.