101. Arbitrary Style Transfer with Style-Attentional Networks
图片风格转换(根据另一幅图片的风格)
102. Reliable and Efficient Image Cropping: A Grid Anchor based Approach
可靠高效的图像裁剪:基于网格锚点的方法
103. Photo Wake-Up Character Animation from a Single Photo
根据一张图像生成图像中人物的3D动画
104. DeepLight: Learning Illumination for Unconstrained Mobile Mixed Reality
对于AR,用来渲染虚拟物体的光照应该和现实场景中的光照相同。这篇文章就是要解决如何确定这个虚拟物体的光照。
?105. A Robust Local Spectral Descriptor for Matching Non-Rigid Shapes with Incompatible Shape Structures
为3D形状构建鲁棒且有区别的局部描述符是许多计算机视觉应用的关键组成部分。
106. StoryGAN: A Sequential Conditional GAN for Story Visualization
故事可视化,即通过生成一系列图片来描述一个故事(输入是一段由多个句子组成的故事)
?107. Unsupervised Learning of Dense Shape Correspondence
108. Bidirectional Learning for Domain Adaptation of Semantic Segmentation
语义分割领域适应的双向学习
在本文中,我们提出了一种新的双向学习框架,用于领域适应分割。 使用双向学习,可以交替地学习图像翻译模型和分割适应模型并相互促进。 此外,我们提出了一种自我监督学习算法来学习更好的分割自适应模型,并反过来改进图像转换模型。 实验表明,我们的方法在分割领域适应性方面优于最先进的方法。 源代码可在https://github.com/liyunsheng13/BDL获得。
?109. Sensitive-Sample Fingerprinting of Deep Neural Networks
110. PointNetLK: Robust & Efficient Point Cloud Registration using PointNet
点云注册是找到对齐点云对的空间变换的过程
如图,找到把红色点云对其到黑色点云的变换过程
111. Occlusion-Net: 2D/3D Occluded Keypoint Localization Using Graph Networks
我们提出了Occlusion-Net1,一个以很大程度上自我监督的方式预测对象被遮挡关键点的2D和3D位置的框架
112. Graphonomy: Universal Human Parsing via Graph Transfer Learning
113. Speech2Face: Learning the Face Behind a Voice
输入:声音片段
输出:根据声音重建出的人脸模型
?114. Mixed Effect Neural Networks (MeNets) with Applications to Gaze Estimation
混合效应神经网络(MeNets)及其在Gaze估计中的应用
115. Bayesian Hierarchial Dynamic Model for Human Action Recognition
人类行为识别
116. MMFace: A Multi-Metric Regression Network for Unconstrained Face Reconstruction
人脸重建(可用于表情驱动)
117. RepNet: Weakly Supervised Training of an Adversarial Reprojection Network for 3D Human Pose Estimation
三维人体姿态估计
118. Learning to Regress 3D Face Shape and Expression from an Image Without 3D Supervision
三维人脸表情重建 ?无3D监督?
119. 3D Human Pose Estimation in Video with Temporal Convolution and Semi-supervised Training
三维人体姿态检测(视频)
120. Listen to the Image
121. Text2Scene: Generating Compositional Scenes from Textual Descriptions
Text2Scene:从文本描述生成合成场景
122. Seamless Scene Segmetation
123. Enhanced Pix2pix Dehazing Network
124. Learning to Synthesize Motion Blur
从两个清晰的图像合成运动模糊图像
125. TraVelGAN: Image-to-image Translation by Transformation VectorLearning Matthew Amodio Smita Krishnaswamy