Matches in SemOpenAlex for { <https://semopenalex.org/work/W4378506961> ?p ?o ?g. }
Showing items 1 to 73 of
73
with 100 items per page.
- W4378506961 abstract "We present Cross3DVG, a novel task for cross-dataset visual grounding in 3D scenes, revealing the limitations of existing 3D visual grounding models using restricted 3D resources and thus easily overfit to a specific 3D dataset. To facilitate Cross3DVG, we have created a large-scale 3D visual grounding dataset containing more than 63k diverse descriptions of 3D objects within 1,380 indoor RGB-D scans from 3RScan with human annotations, paired with the existing 52k descriptions on ScanRefer. We perform Cross3DVG by training a model on the source 3D visual grounding dataset and then evaluating it on the target dataset constructed in different ways (e.g., different sensors, 3D reconstruction methods, and language annotators) without using target labels. We conduct comprehensive experiments using established visual grounding models, as well as a CLIP-based 2D-3D integration method, designed to bridge the gaps between 3D datasets. By performing Cross3DVG tasks, we found that (i) cross-dataset 3D visual grounding has significantly lower performance than learning and evaluation with a single dataset, suggesting much room for improvement in cross-dataset generalization of 3D visual grounding, (ii) better detectors and transformer-based localization modules for 3D grounding are beneficial for enhancing 3D grounding performance and (iii) fusing 2D-3D data using CLIP demonstrates further performance improvements. Our Cross3DVG task will provide a benchmark for developing robust 3D visual grounding models capable of handling diverse 3D scenes while leveraging deep language understanding." @default.
- W4378506961 created "2023-05-27" @default.
- W4378506961 creator A5006630406 @default.
- W4378506961 creator A5031443776 @default.
- W4378506961 creator A5055278147 @default.
- W4378506961 creator A5073732915 @default.
- W4378506961 date "2023-05-23" @default.
- W4378506961 modified "2023-09-30" @default.
- W4378506961 title "Cross3DVG: Baseline and Dataset for Cross-Dataset 3D Visual Grounding on Different RGB-D Scans" @default.
- W4378506961 doi "https://doi.org/10.48550/arxiv.2305.13876" @default.
- W4378506961 hasPublicationYear "2023" @default.
- W4378506961 type Work @default.
- W4378506961 citedByCount "0" @default.
- W4378506961 crossrefType "posted-content" @default.
- W4378506961 hasAuthorship W4378506961A5006630406 @default.
- W4378506961 hasAuthorship W4378506961A5031443776 @default.
- W4378506961 hasAuthorship W4378506961A5055278147 @default.
- W4378506961 hasAuthorship W4378506961A5073732915 @default.
- W4378506961 hasBestOaLocation W43785069611 @default.
- W4378506961 hasConcept C119599485 @default.
- W4378506961 hasConcept C119857082 @default.
- W4378506961 hasConcept C127413603 @default.
- W4378506961 hasConcept C13280743 @default.
- W4378506961 hasConcept C134306372 @default.
- W4378506961 hasConcept C153180895 @default.
- W4378506961 hasConcept C154945302 @default.
- W4378506961 hasConcept C168993435 @default.
- W4378506961 hasConcept C177148314 @default.
- W4378506961 hasConcept C185798385 @default.
- W4378506961 hasConcept C201995342 @default.
- W4378506961 hasConcept C205649164 @default.
- W4378506961 hasConcept C22019652 @default.
- W4378506961 hasConcept C2780451532 @default.
- W4378506961 hasConcept C31972630 @default.
- W4378506961 hasConcept C33923547 @default.
- W4378506961 hasConcept C41008148 @default.
- W4378506961 hasConcept C50644808 @default.
- W4378506961 hasConcept C82990744 @default.
- W4378506961 hasConceptScore W4378506961C119599485 @default.
- W4378506961 hasConceptScore W4378506961C119857082 @default.
- W4378506961 hasConceptScore W4378506961C127413603 @default.
- W4378506961 hasConceptScore W4378506961C13280743 @default.
- W4378506961 hasConceptScore W4378506961C134306372 @default.
- W4378506961 hasConceptScore W4378506961C153180895 @default.
- W4378506961 hasConceptScore W4378506961C154945302 @default.
- W4378506961 hasConceptScore W4378506961C168993435 @default.
- W4378506961 hasConceptScore W4378506961C177148314 @default.
- W4378506961 hasConceptScore W4378506961C185798385 @default.
- W4378506961 hasConceptScore W4378506961C201995342 @default.
- W4378506961 hasConceptScore W4378506961C205649164 @default.
- W4378506961 hasConceptScore W4378506961C22019652 @default.
- W4378506961 hasConceptScore W4378506961C2780451532 @default.
- W4378506961 hasConceptScore W4378506961C31972630 @default.
- W4378506961 hasConceptScore W4378506961C33923547 @default.
- W4378506961 hasConceptScore W4378506961C41008148 @default.
- W4378506961 hasConceptScore W4378506961C50644808 @default.
- W4378506961 hasConceptScore W4378506961C82990744 @default.
- W4378506961 hasLocation W43785069611 @default.
- W4378506961 hasOpenAccess W4378506961 @default.
- W4378506961 hasPrimaryLocation W43785069611 @default.
- W4378506961 hasRelatedWork W2052518016 @default.
- W4378506961 hasRelatedWork W2085956791 @default.
- W4378506961 hasRelatedWork W2283162247 @default.
- W4378506961 hasRelatedWork W2314488738 @default.
- W4378506961 hasRelatedWork W2524507886 @default.
- W4378506961 hasRelatedWork W2989932438 @default.
- W4378506961 hasRelatedWork W3011996705 @default.
- W4378506961 hasRelatedWork W3099765033 @default.
- W4378506961 hasRelatedWork W3175189414 @default.
- W4378506961 hasRelatedWork W4212983513 @default.
- W4378506961 isParatext "false" @default.
- W4378506961 isRetracted "false" @default.
- W4378506961 workType "article" @default.