https://scholars.lib.ntu.edu.tw/handle/123456789/607368
標題: | LayoutTransformer: Scene Layout Generation with Conceptual and Spatial Diversity | 作者: | Yang C.-F Fan W.-C Yang F.-E YU-CHIANG WANG |
關鍵字: | Computer vision;Bounding-box;Co-occurrence relationships;Contextual feature;Implicit relationships;Layout generations;Scene-graphs;Semantic features;Spatial diversity;Spatial informations;Text input;Semantics | 公開日期: | 2021 | 起(迄)頁: | 3731-3740 | 來源出版物: | Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition | 摘要: | When translating text inputs into layouts or images, existing works typically require explicit descriptions of each object in a scene, including their spatial information or the associated relationships. To better exploit the text input, so that implicit objects or relationships can be properly inferred during layout generation, we propose a LayoutTransformer Network (LT-Net) in this paper. Given a scene-graph input, our LT-Net uniquely encodes the semantic features for exploiting their co-occurrences and implicit relationships. This allows one to manipulate conceptually diverse yet plausible layout outputs. Moreover, the decoder of our LT-Net translates the encoded contextual features into bounding boxes with self-supervised relation consistency preserved. By fitting their distributions to Gaussian mixture models, spatially-diverse layouts can be additionally produced by LT-Net. We conduct extensive experiments on the datasets of MS-COCO and Visual Genome, and confirm the effectiveness and plausibility of our LT-Net over recent layout generation models. Codes will be released at LayoutTransformer. ? 2021 IEEE |
URI: | https://www.scopus.com/inward/record.uri?eid=2-s2.0-85123204731&doi=10.1109%2fCVPR46437.2021.00373&partnerID=40&md5=3a79edced3a256def9f8d66bb169be6e https://scholars.lib.ntu.edu.tw/handle/123456789/607368 |
ISSN: | 10636919 | DOI: | 10.1109/CVPR46437.2021.00373 |
顯示於: | 電機工程學系 |
在 IR 系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。