A plug-and-play Gaussian-splatting rendering loss that supervises 3D occupancy predictions through 2D depth and semantic images improves geometric fidelity and surface-sensitive metrics across multiple models and benchmarks.
The lov ´asz-softmax loss: A tractable surrogate for the optimization of the intersection-over-union measure in neural networks
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
GaussRender: Learning 3D Occupancy with Gaussian Rendering
A plug-and-play Gaussian-splatting rendering loss that supervises 3D occupancy predictions through 2D depth and semantic images improves geometric fidelity and surface-sensitive metrics across multiple models and benchmarks.