Matches in SemOpenAlex for { <https://semopenalex.org/work/W2901982528> ?p ?o ?g. }
Showing items 1 to 86 of
86
with 100 items per page.
- W2901982528 abstract "3D point-cloud recognition with PointNet and its variants has received remarkable progress. A missing ingredient, however, is the ability to automatically evaluate point-wise importance w.r.t.! classification performance, which is usually reflected by a saliency map. A saliency map is an important tool as it allows one to perform further processes on point-cloud data. In this paper, we propose a novel way of characterizing critical points and segments to build point-cloud saliency maps. Our method assigns each point a score reflecting its contribution to the model-recognition loss. The saliency map explicitly explains which points are the key for model recognition. Furthermore, aggregations of highly-scored points indicate important segments/subsets in a point-cloud. Our motivation for constructing a saliency map is by point dropping, which is a non-differentiable operator. To overcome this issue, we approximate point-dropping with a differentiable procedure of shifting points towards the cloud centroid. Consequently, each saliency score can be efficiently measured by the corresponding gradient of the loss w.r.t the point under the spherical coordinates. Extensive evaluations on several state-of-the-art point-cloud recognition models, including PointNet, PointNet++ and DGCNN, demonstrate the veracity and generality of our proposed saliency map. Code for experiments is released on url{this https URL}." @default.
- W2901982528 created "2018-12-11" @default.
- W2901982528 creator A5006435999 @default.
- W2901982528 creator A5085245110 @default.
- W2901982528 creator A5087996736 @default.
- W2901982528 creator A5091687257 @default.
- W2901982528 date "2018-11-28" @default.
- W2901982528 modified "2023-09-24" @default.
- W2901982528 title "Learning Saliency Maps for Adversarial Point-Cloud Generation" @default.
- W2901982528 cites W2211722331 @default.
- W2901982528 cites W2560609797 @default.
- W2901982528 cites W2560722161 @default.
- W2901982528 cites W2963121255 @default.
- W2901982528 cites W2963389226 @default.
- W2901982528 cites W2963857521 @default.
- W2901982528 hasPublicationYear "2018" @default.
- W2901982528 type Work @default.
- W2901982528 sameAs 2901982528 @default.
- W2901982528 citedByCount "1" @default.
- W2901982528 countsByYear W29019825282019 @default.
- W2901982528 crossrefType "posted-content" @default.
- W2901982528 hasAuthorship W2901982528A5006435999 @default.
- W2901982528 hasAuthorship W2901982528A5085245110 @default.
- W2901982528 hasAuthorship W2901982528A5087996736 @default.
- W2901982528 hasAuthorship W2901982528A5091687257 @default.
- W2901982528 hasConcept C131979681 @default.
- W2901982528 hasConcept C134306372 @default.
- W2901982528 hasConcept C146599234 @default.
- W2901982528 hasConcept C153180895 @default.
- W2901982528 hasConcept C154945302 @default.
- W2901982528 hasConcept C15744967 @default.
- W2901982528 hasConcept C177264268 @default.
- W2901982528 hasConcept C199360897 @default.
- W2901982528 hasConcept C202615002 @default.
- W2901982528 hasConcept C2524010 @default.
- W2901982528 hasConcept C2776760102 @default.
- W2901982528 hasConcept C2780767217 @default.
- W2901982528 hasConcept C28719098 @default.
- W2901982528 hasConcept C31972630 @default.
- W2901982528 hasConcept C33923547 @default.
- W2901982528 hasConcept C41008148 @default.
- W2901982528 hasConcept C542102704 @default.
- W2901982528 hasConceptScore W2901982528C131979681 @default.
- W2901982528 hasConceptScore W2901982528C134306372 @default.
- W2901982528 hasConceptScore W2901982528C146599234 @default.
- W2901982528 hasConceptScore W2901982528C153180895 @default.
- W2901982528 hasConceptScore W2901982528C154945302 @default.
- W2901982528 hasConceptScore W2901982528C15744967 @default.
- W2901982528 hasConceptScore W2901982528C177264268 @default.
- W2901982528 hasConceptScore W2901982528C199360897 @default.
- W2901982528 hasConceptScore W2901982528C202615002 @default.
- W2901982528 hasConceptScore W2901982528C2524010 @default.
- W2901982528 hasConceptScore W2901982528C2776760102 @default.
- W2901982528 hasConceptScore W2901982528C2780767217 @default.
- W2901982528 hasConceptScore W2901982528C28719098 @default.
- W2901982528 hasConceptScore W2901982528C31972630 @default.
- W2901982528 hasConceptScore W2901982528C33923547 @default.
- W2901982528 hasConceptScore W2901982528C41008148 @default.
- W2901982528 hasConceptScore W2901982528C542102704 @default.
- W2901982528 hasLocation W29019825281 @default.
- W2901982528 hasOpenAccess W2901982528 @default.
- W2901982528 hasPrimaryLocation W29019825281 @default.
- W2901982528 hasRelatedWork W1606881493 @default.
- W2901982528 hasRelatedWork W2117183049 @default.
- W2901982528 hasRelatedWork W2158987471 @default.
- W2901982528 hasRelatedWork W2274528245 @default.
- W2901982528 hasRelatedWork W2380505014 @default.
- W2901982528 hasRelatedWork W2509486428 @default.
- W2901982528 hasRelatedWork W2560609797 @default.
- W2901982528 hasRelatedWork W2803937221 @default.
- W2901982528 hasRelatedWork W2902333684 @default.
- W2901982528 hasRelatedWork W2932629318 @default.
- W2901982528 hasRelatedWork W2947324926 @default.
- W2901982528 hasRelatedWork W2960986959 @default.
- W2901982528 hasRelatedWork W2963829411 @default.
- W2901982528 hasRelatedWork W2975950484 @default.
- W2901982528 hasRelatedWork W2980839814 @default.
- W2901982528 hasRelatedWork W2982361312 @default.
- W2901982528 hasRelatedWork W3032119513 @default.
- W2901982528 hasRelatedWork W3110726256 @default.
- W2901982528 hasRelatedWork W3164058023 @default.
- W2901982528 hasRelatedWork W3205474795 @default.
- W2901982528 isParatext "false" @default.
- W2901982528 isRetracted "false" @default.
- W2901982528 magId "2901982528" @default.
- W2901982528 workType "article" @default.