DC Field | Value | Language |
---|---|---|
dc.contributor.author | Jang, Heeoh | ko |
dc.contributor.author | Kim, Dongkyu | ko |
dc.contributor.author | Ahn, Wonhyuk | ko |
dc.contributor.author | Lee, Heung-Kyu | ko |
dc.date.accessioned | 2019-12-13T09:31:00Z | - |
dc.date.available | 2019-12-13T09:31:00Z | - |
dc.date.created | 2019-11-07 | - |
dc.date.created | 2019-11-07 | - |
dc.date.created | 2019-11-07 | - |
dc.date.issued | 2019-09-28 | - |
dc.identifier.citation | 2nd IEEE International Conference on Information Communication and Signal Processing (ICICSP), pp.495 - 499 | - |
dc.identifier.uri | http://hdl.handle.net/10203/269280 | - |
dc.description.abstract | Recently, object detection has presented superior performance using the deep convolutional neural network (CNN). However, most CNN-based models need the bounding box information of the input image in pairs. To overcome this limitation, we propose the Generative Object Detection which learns with only cropped images that are not in pairs. Our model based on Generative Adversarial Networks (GAN) creates cropped images by making a mask that represents the object region. To achieve this goal, we devise a novel mask mean loss (MML) that helps the GAN be able to estimate the distribution of training data and uses dilated convolution for a wider reception field in the generator. The experimental results show that Generative Object Detection improves the mIoU and accuracy. | - |
dc.language | English | - |
dc.publisher | IEEE | - |
dc.title | Generative Object Detection: Erasing the Boundary via Adversarial Learning with Mask | - |
dc.type | Conference | - |
dc.identifier.wosid | 000516605600100 | - |
dc.identifier.scopusid | 2-s2.0-85078887427 | - |
dc.type.rims | CONF | - |
dc.citation.beginningpage | 495 | - |
dc.citation.endingpage | 499 | - |
dc.citation.publicationname | 2nd IEEE International Conference on Information Communication and Signal Processing (ICICSP) | - |
dc.identifier.conferencecountry | CC | - |
dc.identifier.conferencelocation | Shandong University, Weihai | - |
dc.contributor.localauthor | Lee, Heung-Kyu | - |
dc.contributor.nonIdAuthor | Jang, Heeoh | - |
dc.contributor.nonIdAuthor | Kim, Dongkyu | - |
dc.contributor.nonIdAuthor | Ahn, Wonhyuk | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.