Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385804920> ?p ?o ?g. }
Showing items 1 to 70 of
70
with 100 items per page.
- W4385804920 abstract "In the realm of 3D-computer vision applications, point cloud few-shot learning plays a critical role. However, it poses an arduous challenge due to the sparsity, irregularity, and unordered nature of the data. Current methods rely on complex local geometric extraction techniques such as convolution, graph, and attention mechanisms, along with extensive data-driven pre-training tasks. These approaches contradict the fundamental goal of few-shot learning, which is to facilitate efficient learning. To address this issue, we propose GPr-Net (Geometric Prototypical Network), a lightweight and computationally efficient geometric prototypical network that captures the intrinsic topology of point clouds and achieves superior performance. Our proposed method, IGI++ (Intrinsic Geometry Interpreter++) employs vector-based hand-crafted intrinsic geometry interpreters and Laplace vectors to extract and evaluate point cloud morphology, resulting in improved representations for FSL (Few-Shot Learning). Additionally, Laplace vectors enable the extraction of valuable features from point clouds with fewer points. To tackle the distribution drift challenge in few-shot metric learning, we leverage hyperbolic space and demonstrate that our approach handles intra and inter-class variance better than existing point cloud few-shot learning methods. Experimental results on the ModelNet40 dataset show that GPr-Net outperforms state-of-the-art methods in few-shot learning on point clouds, achieving utmost computational efficiency that is 170× better than all existing works. The code is publicly available at https://github.com/TejasAnvekar/GPr-Net." @default.
- W4385804920 created "2023-08-15" @default.
- W4385804920 creator A5028590022 @default.
- W4385804920 creator A5083531509 @default.
- W4385804920 date "2023-06-01" @default.
- W4385804920 modified "2023-09-26" @default.
- W4385804920 title "GPr-Net: Geometric Prototypical Network for Point Cloud Few-Shot Learning" @default.
- W4385804920 cites W1644641054 @default.
- W4385804920 cites W1989625560 @default.
- W4385804920 cites W2072723786 @default.
- W4385804920 cites W2098764590 @default.
- W4385804920 cites W2162112538 @default.
- W4385804920 cites W2889326414 @default.
- W4385804920 cites W2962887844 @default.
- W4385804920 cites W2964228567 @default.
- W4385804920 cites W2979750740 @default.
- W4385804920 cites W3035363555 @default.
- W4385804920 cites W3096063621 @default.
- W4385804920 cites W3104240813 @default.
- W4385804920 cites W3132231790 @default.
- W4385804920 cites W3168377947 @default.
- W4385804920 cites W3210638576 @default.
- W4385804920 cites W4213284407 @default.
- W4385804920 cites W4224917275 @default.
- W4385804920 cites W4226044115 @default.
- W4385804920 cites W4292793987 @default.
- W4385804920 cites W4294770622 @default.
- W4385804920 cites W4318624522 @default.
- W4385804920 cites W4319299876 @default.
- W4385804920 doi "https://doi.org/10.1109/cvprw59228.2023.00440" @default.
- W4385804920 hasPublicationYear "2023" @default.
- W4385804920 type Work @default.
- W4385804920 citedByCount "0" @default.
- W4385804920 crossrefType "proceedings-article" @default.
- W4385804920 hasAuthorship W4385804920A5028590022 @default.
- W4385804920 hasAuthorship W4385804920A5083531509 @default.
- W4385804920 hasBestOaLocation W43858049202 @default.
- W4385804920 hasConcept C111919701 @default.
- W4385804920 hasConcept C11413529 @default.
- W4385804920 hasConcept C131979681 @default.
- W4385804920 hasConcept C153083717 @default.
- W4385804920 hasConcept C154945302 @default.
- W4385804920 hasConcept C31972630 @default.
- W4385804920 hasConcept C41008148 @default.
- W4385804920 hasConcept C79974875 @default.
- W4385804920 hasConceptScore W4385804920C111919701 @default.
- W4385804920 hasConceptScore W4385804920C11413529 @default.
- W4385804920 hasConceptScore W4385804920C131979681 @default.
- W4385804920 hasConceptScore W4385804920C153083717 @default.
- W4385804920 hasConceptScore W4385804920C154945302 @default.
- W4385804920 hasConceptScore W4385804920C31972630 @default.
- W4385804920 hasConceptScore W4385804920C41008148 @default.
- W4385804920 hasConceptScore W4385804920C79974875 @default.
- W4385804920 hasLocation W43858049201 @default.
- W4385804920 hasLocation W43858049202 @default.
- W4385804920 hasOpenAccess W4385804920 @default.
- W4385804920 hasPrimaryLocation W43858049201 @default.
- W4385804920 hasRelatedWork W2033574012 @default.
- W4385804920 hasRelatedWork W2383532021 @default.
- W4385804920 hasRelatedWork W2979718872 @default.
- W4385804920 hasRelatedWork W3046762217 @default.
- W4385804920 hasRelatedWork W3128716822 @default.
- W4385804920 hasRelatedWork W3158534694 @default.
- W4385804920 hasRelatedWork W3206828132 @default.
- W4385804920 hasRelatedWork W4287694812 @default.
- W4385804920 hasRelatedWork W4290774832 @default.
- W4385804920 hasRelatedWork W4293067784 @default.
- W4385804920 isParatext "false" @default.
- W4385804920 isRetracted "false" @default.
- W4385804920 workType "article" @default.