Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386075863> ?p ?o ?g. }
- W4386075863 abstract "Generalizable 3D part segmentation is important but challenging in vision and robotics. Training deep models via conventional supervised methods requires large-scale 3D datasets with fine-grained part annotations, which are costly to collect. This paper explores an alternative way for low-shot part segmentation of 3D point clouds by leveraging a pretrained image-language model, GLIP. which achieves superior performance on open-vocabulary 2D detection. We transfer the rich knowledge from 2D to 3D through GLIP-based part detection on point cloud rendering and a novel 2D-to-3D label lifting algorithm. We also utilize multi-view 3D priors and few-shot prompt tuning to boost performance significantly. Extensive evaluation on PartNet and PartNet-Mobility datasets shows that our method enables excellent zero-shot 3D part segmentation. Our few-shot version not only outperforms existing few-shot approaches by a large margin but also achieves highly competitive results compared to the fully supervised counterpart. Furthermore, we demonstrate that our method can be directly applied to iPhone-scanned point clouds without significant domain gaps." @default.
- W4386075863 created "2023-08-23" @default.
- W4386075863 creator A5006266154 @default.
- W4386075863 creator A5013879989 @default.
- W4386075863 creator A5019886822 @default.
- W4386075863 creator A5038550641 @default.
- W4386075863 creator A5053652981 @default.
- W4386075863 creator A5058453644 @default.
- W4386075863 creator A5059083339 @default.
- W4386075863 date "2023-06-01" @default.
- W4386075863 modified "2023-09-29" @default.
- W4386075863 title "PartSLIP: Low-Shot Part Segmentation for 3D Point Clouds via Pretrained Image-Language Models" @default.
- W4386075863 cites W1974537989 @default.
- W4386075863 cites W2346698422 @default.
- W4386075863 cites W2553307952 @default.
- W4386075863 cites W2769312834 @default.
- W4386075863 cites W2904332125 @default.
- W4386075863 cites W2918473435 @default.
- W4386075863 cites W2959771705 @default.
- W4386075863 cites W2961368225 @default.
- W4386075863 cites W2962759414 @default.
- W4386075863 cites W2963281829 @default.
- W4386075863 cites W2963509914 @default.
- W4386075863 cites W2979750740 @default.
- W4386075863 cites W2989341556 @default.
- W4386075863 cites W2990613095 @default.
- W4386075863 cites W2991485494 @default.
- W4386075863 cites W3034949383 @default.
- W4386075863 cites W3035624836 @default.
- W4386075863 cites W3035739565 @default.
- W4386075863 cites W3109051808 @default.
- W4386075863 cites W3110047846 @default.
- W4386075863 cites W3118038951 @default.
- W4386075863 cites W3161309385 @default.
- W4386075863 cites W3166573884 @default.
- W4386075863 cites W3168377947 @default.
- W4386075863 cites W3169120894 @default.
- W4386075863 cites W3170500059 @default.
- W4386075863 cites W3177450673 @default.
- W4386075863 cites W3179682050 @default.
- W4386075863 cites W3182524856 @default.
- W4386075863 cites W3196869447 @default.
- W4386075863 cites W3201927751 @default.
- W4386075863 cites W3202995573 @default.
- W4386075863 cites W3203495808 @default.
- W4386075863 cites W3210613940 @default.
- W4386075863 cites W3216343037 @default.
- W4386075863 cites W4212956978 @default.
- W4386075863 cites W4226058394 @default.
- W4386075863 cites W4229456580 @default.
- W4386075863 cites W4283221338 @default.
- W4386075863 cites W4285123658 @default.
- W4386075863 cites W4286611314 @default.
- W4386075863 cites W4312249789 @default.
- W4386075863 cites W4312274934 @default.
- W4386075863 cites W4312674065 @default.
- W4386075863 cites W4312708649 @default.
- W4386075863 cites W4312803310 @default.
- W4386075863 cites W4312818263 @default.
- W4386075863 cites W4312956471 @default.
- W4386075863 cites W4312971576 @default.
- W4386075863 cites W4313041309 @default.
- W4386075863 cites W4313118252 @default.
- W4386075863 cites W4313169553 @default.
- W4386075863 doi "https://doi.org/10.1109/cvpr52729.2023.02082" @default.
- W4386075863 hasPublicationYear "2023" @default.
- W4386075863 type Work @default.
- W4386075863 citedByCount "1" @default.
- W4386075863 countsByYear W43860758632023 @default.
- W4386075863 crossrefType "proceedings-article" @default.
- W4386075863 hasAuthorship W4386075863A5006266154 @default.
- W4386075863 hasAuthorship W4386075863A5013879989 @default.
- W4386075863 hasAuthorship W4386075863A5019886822 @default.
- W4386075863 hasAuthorship W4386075863A5038550641 @default.
- W4386075863 hasAuthorship W4386075863A5053652981 @default.
- W4386075863 hasAuthorship W4386075863A5058453644 @default.
- W4386075863 hasAuthorship W4386075863A5059083339 @default.
- W4386075863 hasConcept C119857082 @default.
- W4386075863 hasConcept C124504099 @default.
- W4386075863 hasConcept C127413603 @default.
- W4386075863 hasConcept C131979681 @default.
- W4386075863 hasConcept C138885662 @default.
- W4386075863 hasConcept C153180895 @default.
- W4386075863 hasConcept C154945302 @default.
- W4386075863 hasConcept C178790620 @default.
- W4386075863 hasConcept C185592680 @default.
- W4386075863 hasConcept C205711294 @default.
- W4386075863 hasConcept C2524010 @default.
- W4386075863 hasConcept C2777601683 @default.
- W4386075863 hasConcept C2778344882 @default.
- W4386075863 hasConcept C28719098 @default.
- W4386075863 hasConcept C2992734406 @default.
- W4386075863 hasConcept C31972630 @default.
- W4386075863 hasConcept C33923547 @default.
- W4386075863 hasConcept C41008148 @default.
- W4386075863 hasConcept C41895202 @default.
- W4386075863 hasConcept C774472 @default.
- W4386075863 hasConcept C78519656 @default.
- W4386075863 hasConcept C89600930 @default.
- W4386075863 hasConceptScore W4386075863C119857082 @default.