
Paper Notes
文章平均质量分 94
Read paper every day.
zcx_language
这个作者很懒,什么都没留下…
展开
专栏收录文章
- 默认排序
- 最新发布
- 最早发布
- 最多阅读
- 最少阅读
-
Face Forgery Suvery
Face Forgery Survey原创 2023-03-11 18:12:41 · 608 阅读 · 0 评论 -
Local Feature Detection and Description
IntroductionSuperPoint: Self-supervise Interest Point Detection and DescriptionAbstractMagicPoint利用简单几何图像的关键点易于提取且定义较为明确的特点来构造简单数据集,并基于此训练简单几何图像兴趣点检测模型MagicPoint。Homographic Adaptation作者利用MS-COCO数据集和上述训练的MagicPoint模型,并结合其提出的Homographic Adaptation机原创 2021-04-22 19:33:48 · 643 阅读 · 0 评论 -
Multi-Label Image Recognition with Graph Convolutional Networks(CVPR 2019)
Large-Scale Object Detection in the Wild from Imbalanced Multi-LabelsPaper PDF文章目录IntorductionInnovationMethodExperimentIntorductionInnovationMethodExperiment原创 2020-06-18 12:59:26 · 931 阅读 · 1 评论 -
CNN-RNN: A Unified Framework for Multi-label Image Classification(CVPR 2016)
CNN-RNN: A Unified Framework for Multi-label Image ClassificationPaper PDF文章目录IntroductInnovationMethodExperimentIntroductInnovationMethodExperiment原创 2020-06-15 18:21:32 · 1052 阅读 · 1 评论 -
Learning Generalized Spoof Cues for Face Anti-spoofing
Learning Generalized Spoof Cues for Face Anti-spoofingpaper PDFhttps://blog.youkuaiyun.com/fuwenyan/article/details/77769224原创 2020-06-01 12:37:11 · 2485 阅读 · 0 评论 -
Seeing Voices and Hearing Faces: Cross-modal biometric matching(CVPR 2018)
Seeing Voices and Hearing Faces: Cross-modal biometric matchingPaper PDF文章目录IntroductionInnovationMethodExperimentIntroductionInnovationMethodExperiment原创 2020-05-28 14:28:03 · 834 阅读 · 0 评论 -
Self-supervised Representation Learning from Videos for Facial Action Unit Detection(CVPR 2019)
Self-supervised Representation Learning from Videos for Facial Action Unit Detection(CVPR 2019)Paper PDF文章目录IntroductionInnovationMethodFeature disentanglingTarget reconstructionCycle with AU/pose changedExperimentImplementation details DetailedCompariso原创 2020-05-24 18:41:24 · 1702 阅读 · 0 评论 -
Detecting Deep-Fake Videos from Appearance and Behavior
Face-swap方式,将target的脸替换为source的脸,形成deepfake video.这种方式只是改变了target的型,而没有改变target的神,意思就是说deepfake video中人物的神情习惯仍然是target的。作者利用这一特征。通过将人物identity和behavior feature关联起来。作者分别以以video中的behavior feature和face为线...原创 2020-05-20 12:36:56 · 2270 阅读 · 0 评论 -
Orderless Recurrent Models for Multi-label Classification (CVPR2020)
Orderless Recurrent Models for Multi-label Classification Paper PDF文章目录IntroductionInnovationMethodImage-to-sequence modelTraining recurrent modelsOrderless recurrent modelsExperimentsIntroductionM...原创 2020-05-06 17:07:39 · 1136 阅读 · 0 评论 -
WS-DAN:Weakly Supervised Data Augmentation Netowrk for Fine-Grained Visual Classification
See Better Before Looking Closer: Weakly Supervised Data Augmentation Netowrk for Fine-Grained Visual ClassificationPaper PDF文章目录AbstractPipelineWeakly Supervised Attention LearningSpatial Represent...原创 2020-05-03 18:30:00 · 779 阅读 · 0 评论 -
Mask RCNN
文章目录Noun interpretationRCNNRPNFeature Map上采样下采样高斯金字塔、高斯平滑、亚采样bottom-up网络top-down网络Noun interpretationRCNNRPNFeature Pyramid NetworksFeature Map在每个卷积层,数据都是以三维形式存在的。你可以把它看成许多个二维图片叠在一起,其中每一个称为一个fea...原创 2019-10-20 16:35:44 · 669 阅读 · 0 评论