Matches in SemOpenAlex for { <https://semopenalex.org/work/W4304080269> ?p ?o ?g. }
- W4304080269 abstract "We present PVSeRF, a learning framework that reconstructs neural radiance fields from single-view RGB images, for novel view synthesis. Previous solutions, such as pixelNeRF, rely only on pixel-aligned features and suffer from feature ambiguity issues. As a result, they struggle with the disentanglement of geometry and appearance, leading to implausible geometries and blurry results. To address this challenge, we propose to incorporate explicit geometry reasoning and combine it with pixel-aligned features for radiance field prediction. Specifically, in addition to pixel-aligned features, we further constrain the radiance field learning to be conditioned on i) voxel-aligned features learned from a coarse volumetric grid and ii) fine surface-aligned features extracted from a regressed point cloud. We show that the introduction of such geometry-aware features helps to achieve a better disentanglement between appearance and geometry, i.e. recovering more accurate geometries and synthesizing higher quality images of novel views. Extensive experiments against state-of-the-art methods on ShapeNet benchmarks demonstrate the superiority of our approach for single-image novel view synthesis." @default.
- W4304080269 created "2022-10-10" @default.
- W4304080269 creator A5003710372 @default.
- W4304080269 creator A5009164482 @default.
- W4304080269 creator A5015737161 @default.
- W4304080269 creator A5034416956 @default.
- W4304080269 creator A5042771880 @default.
- W4304080269 creator A5065706581 @default.
- W4304080269 creator A5079937813 @default.
- W4304080269 date "2022-10-10" @default.
- W4304080269 modified "2023-09-27" @default.
- W4304080269 title "PVSeRF: Joint Pixel-, Voxel- and Surface-Aligned Radiance Field for Single-Image Novel View Synthesis" @default.
- W4304080269 cites W1539230104 @default.
- W4304080269 cites W2063366997 @default.
- W4304080269 cites W2117248802 @default.
- W4304080269 cites W2122572959 @default.
- W4304080269 cites W2128052895 @default.
- W4304080269 cites W2133665775 @default.
- W4304080269 cites W2138011018 @default.
- W4304080269 cites W2194775991 @default.
- W4304080269 cites W2294985758 @default.
- W4304080269 cites W2348664362 @default.
- W4304080269 cites W2495603374 @default.
- W4304080269 cites W2598591334 @default.
- W4304080269 cites W2796426482 @default.
- W4304080269 cites W2902266071 @default.
- W4304080269 cites W2905288042 @default.
- W4304080269 cites W2933283236 @default.
- W4304080269 cites W2961368225 @default.
- W4304080269 cites W2962785568 @default.
- W4304080269 cites W2962849139 @default.
- W4304080269 cites W2963527086 @default.
- W4304080269 cites W2963926543 @default.
- W4304080269 cites W2963942118 @default.
- W4304080269 cites W2968257580 @default.
- W4304080269 cites W2993195924 @default.
- W4304080269 cites W2997343139 @default.
- W4304080269 cites W3034801905 @default.
- W4304080269 cites W3034968345 @default.
- W4304080269 cites W3035507572 @default.
- W4304080269 cites W3035524985 @default.
- W4304080269 cites W3105863736 @default.
- W4304080269 cites W3109585842 @default.
- W4304080269 cites W3117476483 @default.
- W4304080269 cites W3171145773 @default.
- W4304080269 cites W3172568571 @default.
- W4304080269 cites W3176368002 @default.
- W4304080269 cites W3186630079 @default.
- W4304080269 cites W3197617697 @default.
- W4304080269 cites W3211538617 @default.
- W4304080269 cites W4214731463 @default.
- W4304080269 doi "https://doi.org/10.1145/3503161.3547893" @default.
- W4304080269 hasPublicationYear "2022" @default.
- W4304080269 type Work @default.
- W4304080269 citedByCount "2" @default.
- W4304080269 countsByYear W43040802692022 @default.
- W4304080269 countsByYear W43040802692023 @default.
- W4304080269 crossrefType "proceedings-article" @default.
- W4304080269 hasAuthorship W4304080269A5003710372 @default.
- W4304080269 hasAuthorship W4304080269A5009164482 @default.
- W4304080269 hasAuthorship W4304080269A5015737161 @default.
- W4304080269 hasAuthorship W4304080269A5034416956 @default.
- W4304080269 hasAuthorship W4304080269A5042771880 @default.
- W4304080269 hasAuthorship W4304080269A5065706581 @default.
- W4304080269 hasAuthorship W4304080269A5079937813 @default.
- W4304080269 hasBestOaLocation W43040802692 @default.
- W4304080269 hasConcept C127313418 @default.
- W4304080269 hasConcept C131979681 @default.
- W4304080269 hasConcept C138885662 @default.
- W4304080269 hasConcept C154945302 @default.
- W4304080269 hasConcept C160633673 @default.
- W4304080269 hasConcept C187691185 @default.
- W4304080269 hasConcept C199360897 @default.
- W4304080269 hasConcept C202444582 @default.
- W4304080269 hasConcept C205711294 @default.
- W4304080269 hasConcept C23690007 @default.
- W4304080269 hasConcept C2524010 @default.
- W4304080269 hasConcept C2776401178 @default.
- W4304080269 hasConcept C2776449333 @default.
- W4304080269 hasConcept C2776799497 @default.
- W4304080269 hasConcept C2780522230 @default.
- W4304080269 hasConcept C28719098 @default.
- W4304080269 hasConcept C31972630 @default.
- W4304080269 hasConcept C33923547 @default.
- W4304080269 hasConcept C41008148 @default.
- W4304080269 hasConcept C41895202 @default.
- W4304080269 hasConcept C54170458 @default.
- W4304080269 hasConcept C62649853 @default.
- W4304080269 hasConcept C82990744 @default.
- W4304080269 hasConcept C9652623 @default.
- W4304080269 hasConceptScore W4304080269C127313418 @default.
- W4304080269 hasConceptScore W4304080269C131979681 @default.
- W4304080269 hasConceptScore W4304080269C138885662 @default.
- W4304080269 hasConceptScore W4304080269C154945302 @default.
- W4304080269 hasConceptScore W4304080269C160633673 @default.
- W4304080269 hasConceptScore W4304080269C187691185 @default.
- W4304080269 hasConceptScore W4304080269C199360897 @default.
- W4304080269 hasConceptScore W4304080269C202444582 @default.
- W4304080269 hasConceptScore W4304080269C205711294 @default.
- W4304080269 hasConceptScore W4304080269C23690007 @default.