Matches in SemOpenAlex for { <https://semopenalex.org/work/W4310921653> ?p ?o ?g. }
Showing items 1 to 62 of
62
with 100 items per page.
- W4310921653 abstract "2D-to-3D reconstruction is an ill-posed problem, yet humans are good at solving this problem due to their prior knowledge of the 3D world developed over years. Driven by this observation, we propose NeRDi, a single-view NeRF synthesis framework with general image priors from 2D diffusion models. Formulating single-view reconstruction as an image-conditioned 3D generation problem, we optimize the NeRF representations by minimizing a diffusion loss on its arbitrary view renderings with a pretrained image diffusion model under the input-view constraint. We leverage off-the-shelf vision-language models and introduce a two-section language guidance as conditioning inputs to the diffusion model. This is essentially helpful for improving multiview content coherence as it narrows down the general image prior conditioned on the semantic and visual features of the single-view input image. Additionally, we introduce a geometric loss based on estimated depth maps to regularize the underlying 3D geometry of the NeRF. Experimental results on the DTU MVS dataset show that our method can synthesize novel views with higher quality even compared to existing methods trained on this dataset. We also demonstrate our generalizability in zero-shot NeRF synthesis for in-the-wild images." @default.
- W4310921653 created "2022-12-21" @default.
- W4310921653 creator A5023597320 @default.
- W4310921653 creator A5027357870 @default.
- W4310921653 creator A5050560702 @default.
- W4310921653 creator A5052854215 @default.
- W4310921653 creator A5058263282 @default.
- W4310921653 creator A5065368881 @default.
- W4310921653 creator A5081024054 @default.
- W4310921653 date "2022-12-06" @default.
- W4310921653 modified "2023-10-16" @default.
- W4310921653 title "NeRDi: Single-View NeRF Synthesis with Language-Guided Diffusion as General Image Priors" @default.
- W4310921653 doi "https://doi.org/10.48550/arxiv.2212.03267" @default.
- W4310921653 hasPublicationYear "2022" @default.
- W4310921653 type Work @default.
- W4310921653 citedByCount "0" @default.
- W4310921653 crossrefType "posted-content" @default.
- W4310921653 hasAuthorship W4310921653A5023597320 @default.
- W4310921653 hasAuthorship W4310921653A5027357870 @default.
- W4310921653 hasAuthorship W4310921653A5050560702 @default.
- W4310921653 hasAuthorship W4310921653A5052854215 @default.
- W4310921653 hasAuthorship W4310921653A5058263282 @default.
- W4310921653 hasAuthorship W4310921653A5065368881 @default.
- W4310921653 hasAuthorship W4310921653A5081024054 @default.
- W4310921653 hasBestOaLocation W43109216531 @default.
- W4310921653 hasConcept C107673813 @default.
- W4310921653 hasConcept C115961682 @default.
- W4310921653 hasConcept C153083717 @default.
- W4310921653 hasConcept C154945302 @default.
- W4310921653 hasConcept C177769412 @default.
- W4310921653 hasConcept C2524010 @default.
- W4310921653 hasConcept C2776036281 @default.
- W4310921653 hasConcept C31972630 @default.
- W4310921653 hasConcept C33923547 @default.
- W4310921653 hasConcept C41008148 @default.
- W4310921653 hasConceptScore W4310921653C107673813 @default.
- W4310921653 hasConceptScore W4310921653C115961682 @default.
- W4310921653 hasConceptScore W4310921653C153083717 @default.
- W4310921653 hasConceptScore W4310921653C154945302 @default.
- W4310921653 hasConceptScore W4310921653C177769412 @default.
- W4310921653 hasConceptScore W4310921653C2524010 @default.
- W4310921653 hasConceptScore W4310921653C2776036281 @default.
- W4310921653 hasConceptScore W4310921653C31972630 @default.
- W4310921653 hasConceptScore W4310921653C33923547 @default.
- W4310921653 hasConceptScore W4310921653C41008148 @default.
- W4310921653 hasLocation W43109216531 @default.
- W4310921653 hasLocation W43109216532 @default.
- W4310921653 hasOpenAccess W4310921653 @default.
- W4310921653 hasPrimaryLocation W43109216531 @default.
- W4310921653 hasRelatedWork W2029790457 @default.
- W4310921653 hasRelatedWork W2085259108 @default.
- W4310921653 hasRelatedWork W2380007170 @default.
- W4310921653 hasRelatedWork W2580650124 @default.
- W4310921653 hasRelatedWork W2968424575 @default.
- W4310921653 hasRelatedWork W3041320102 @default.
- W4310921653 hasRelatedWork W3122088529 @default.
- W4310921653 hasRelatedWork W3142333283 @default.
- W4310921653 hasRelatedWork W3181673064 @default.
- W4310921653 hasRelatedWork W4386190339 @default.
- W4310921653 isParatext "false" @default.
- W4310921653 isRetracted "false" @default.
- W4310921653 workType "article" @default.