Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386071569> ?p ?o ?g. }
Showing items 1 to 96 of
96
with 100 items per page.
- W4386071569 abstract "Spherical image object detection emerges in many applications from virtual reality to robotics and automatic driving, while many existing detectors use <tex xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>$l_{n}$</tex> -norms loss for regression of spherical bounding boxes. There are two intrinsic flaws for <tex xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>$l_{n}$</tex> -norms loss, i.e., independent optimization of parameters and inconsistency between metric (dominated by IoU) and loss. These problems are common in planar image detection but more significant in spherical image detection. Solution for these problems has been extensively discussed in planar image detection by using IoU loss and related variants. However, these solutions cannot be migrated to spherical image object detection due to the undifferentiable of the Spherical IoU (SphIoU). In this paper, we design a simple but effective regression loss based on Gaussian Label Distribution Learning (GLDL) for spherical image object detection. Besides, we observe that the scale of the object in a spherical image varies greatly. The huge differences among objects from different categories make the sample selection strategy based on SphIoU challenging. Therefore, we propose GLDL-ATSS as a better training sample selection strategy for objects of the spherical image, which can alleviate the drawback of IoU threshold-based strategy of scale-sample imbalance. Extensive results on various two datasets with different baseline detectors show the effectiveness of our approach." @default.
- W4386071569 created "2023-08-23" @default.
- W4386071569 creator A5028926076 @default.
- W4386071569 creator A5031604002 @default.
- W4386071569 creator A5053250613 @default.
- W4386071569 creator A5054311881 @default.
- W4386071569 creator A5063969891 @default.
- W4386071569 creator A5066768790 @default.
- W4386071569 date "2023-06-01" @default.
- W4386071569 modified "2023-09-27" @default.
- W4386071569 title "Gaussian Label Distribution Learning for Spherical Image Object Detection" @default.
- W4386071569 cites W2194775991 @default.
- W4386071569 cites W2565639579 @default.
- W4386071569 cites W2623181870 @default.
- W4386071569 cites W2795473781 @default.
- W4386071569 cites W2809453856 @default.
- W4386071569 cites W2962766617 @default.
- W4386071569 cites W2963339238 @default.
- W4386071569 cites W2963351448 @default.
- W4386071569 cites W2964306713 @default.
- W4386071569 cites W2979485822 @default.
- W4386071569 cites W2997462482 @default.
- W4386071569 cites W2997747012 @default.
- W4386071569 cites W3004734937 @default.
- W4386071569 cites W3035396860 @default.
- W4386071569 cites W3098090606 @default.
- W4386071569 cites W3130316726 @default.
- W4386071569 cites W3138516171 @default.
- W4386071569 cites W3175496347 @default.
- W4386071569 cites W3194110717 @default.
- W4386071569 cites W4312478570 @default.
- W4386071569 cites W639708223 @default.
- W4386071569 doi "https://doi.org/10.1109/cvpr52729.2023.00106" @default.
- W4386071569 hasPublicationYear "2023" @default.
- W4386071569 type Work @default.
- W4386071569 citedByCount "0" @default.
- W4386071569 crossrefType "proceedings-article" @default.
- W4386071569 hasAuthorship W4386071569A5028926076 @default.
- W4386071569 hasAuthorship W4386071569A5031604002 @default.
- W4386071569 hasAuthorship W4386071569A5053250613 @default.
- W4386071569 hasAuthorship W4386071569A5054311881 @default.
- W4386071569 hasAuthorship W4386071569A5063969891 @default.
- W4386071569 hasAuthorship W4386071569A5066768790 @default.
- W4386071569 hasConcept C115961682 @default.
- W4386071569 hasConcept C119857082 @default.
- W4386071569 hasConcept C121332964 @default.
- W4386071569 hasConcept C153180895 @default.
- W4386071569 hasConcept C154945302 @default.
- W4386071569 hasConcept C162324750 @default.
- W4386071569 hasConcept C163716315 @default.
- W4386071569 hasConcept C176217482 @default.
- W4386071569 hasConcept C198531522 @default.
- W4386071569 hasConcept C21547014 @default.
- W4386071569 hasConcept C2776151529 @default.
- W4386071569 hasConcept C2781238097 @default.
- W4386071569 hasConcept C31972630 @default.
- W4386071569 hasConcept C41008148 @default.
- W4386071569 hasConcept C62520636 @default.
- W4386071569 hasConcept C63584917 @default.
- W4386071569 hasConcept C81692654 @default.
- W4386071569 hasConcept C97355855 @default.
- W4386071569 hasConceptScore W4386071569C115961682 @default.
- W4386071569 hasConceptScore W4386071569C119857082 @default.
- W4386071569 hasConceptScore W4386071569C121332964 @default.
- W4386071569 hasConceptScore W4386071569C153180895 @default.
- W4386071569 hasConceptScore W4386071569C154945302 @default.
- W4386071569 hasConceptScore W4386071569C162324750 @default.
- W4386071569 hasConceptScore W4386071569C163716315 @default.
- W4386071569 hasConceptScore W4386071569C176217482 @default.
- W4386071569 hasConceptScore W4386071569C198531522 @default.
- W4386071569 hasConceptScore W4386071569C21547014 @default.
- W4386071569 hasConceptScore W4386071569C2776151529 @default.
- W4386071569 hasConceptScore W4386071569C2781238097 @default.
- W4386071569 hasConceptScore W4386071569C31972630 @default.
- W4386071569 hasConceptScore W4386071569C41008148 @default.
- W4386071569 hasConceptScore W4386071569C62520636 @default.
- W4386071569 hasConceptScore W4386071569C63584917 @default.
- W4386071569 hasConceptScore W4386071569C81692654 @default.
- W4386071569 hasConceptScore W4386071569C97355855 @default.
- W4386071569 hasFunder F4320321001 @default.
- W4386071569 hasLocation W43860715691 @default.
- W4386071569 hasOpenAccess W4386071569 @default.
- W4386071569 hasPrimaryLocation W43860715691 @default.
- W4386071569 hasRelatedWork W1837097281 @default.
- W4386071569 hasRelatedWork W1966410754 @default.
- W4386071569 hasRelatedWork W1988485990 @default.
- W4386071569 hasRelatedWork W2007544051 @default.
- W4386071569 hasRelatedWork W2095705906 @default.
- W4386071569 hasRelatedWork W2334336442 @default.
- W4386071569 hasRelatedWork W2732308154 @default.
- W4386071569 hasRelatedWork W2801801420 @default.
- W4386071569 hasRelatedWork W2975200075 @default.
- W4386071569 hasRelatedWork W3177406559 @default.
- W4386071569 isParatext "false" @default.
- W4386071569 isRetracted "false" @default.
- W4386071569 workType "article" @default.