Depth Anything V2 delivers finer, more robust monocular depth predictions by replacing real labeled images with synthetic data, scaling the teacher model, and using large-scale pseudo-labeled real images for student training.
Inserf: text-driven generative object insertion in neural 3d scenes
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 2verdicts
UNVERDICTED 2roles
background 1polarities
background 1representative citing papers
D3DR optimizes inserted 3DGS objects with a DDS-inspired diffusion objective plus a new personalization step to match scene lighting, reporting 2 dB PSNR gain over prior methods.
citing papers explorer
-
Depth Anything V2
Depth Anything V2 delivers finer, more robust monocular depth predictions by replacing real labeled images with synthetic data, scaling the teacher model, and using large-scale pseudo-labeled real images for student training.
-
Diffusion Models are Secretly Zero-Shot 3DGS Harmonizers
D3DR optimizes inserted 3DGS objects with a DDS-inspired diffusion objective plus a new personalization step to match scene lighting, reporting 2 dB PSNR gain over prior methods.