报告论文安排

时间：2019年1月13日（周日）8:30-12:30，13:30-17:30

地点：电3楼632会议室

  1. 每篇论文报告时间为12分钟左右（包括展示video，建议ppt不超过15页），提问3分钟左右。
  2. 由于每篇论文的报告时间有限，比较短，因此报告的重点在于：简单介绍作者和单位情况； 主要讲一下这篇文章在做什么(What)？为什么做这个问题(motivation-Why)？这个问题以前有些什么工作？ 难点在哪里？这篇文章解决了什么样的难点或提出什么样的新方法(selling points)？局限和未来工作(limitation and future work)？ 细节若太多不用展开，关键要让大家知道这篇文章的主要思想和价值所在。
  3. 请大家尽量在网上去找关于你所负责报告论文的相关资料。如各个作者的主页，项目网页。 很多作者会把论文、video、甚至将PPT放在他的主页。讲解论文时，论文中的图片、视频均播放给大家看一下。 “一图胜千言”，视频又能胜于图片。
  4. 书面报告要求采用科研论文格式（标题、摘要、关键字、正文、参考文献等）撰写， 内容可以是所报告论文的概览、同主题多篇论文的综述等。
  5. 评分：PPT口头报告 + 书面报告
学号姓名论文
SF18011001Synthetic depth-of-field with a single-camera mobile phone
SF18011002Headon: real-time reenactment of human portrait videos
SF18011003Denoising with kernel prediction and asymmetric loss functions
SF18011004Toonsynth: example-based synthesis of hand-colored cartoon animations
SF18011005Deep unsupervised pixelization
SF18011006Practical dynamic facial appearance modeling and acquisition
SF18011007Stabilized real-time face tracking via a learned dynamic rigidity prior
SF18011008Deep appearance models for face rendering
SF18011009Image smoothing via unsupervised learning
SF18011010Deep incremental learning for efficient high-fidelity face tracking
SF18011011Instant 3D photography
SF18011012Stereo magnification: learning view synthesis using multiplane images
SF18011014Efficient palette-based decomposition and recoloring of images via RGBXY-space geometry
SF18011015Faceshop: deep sketch-based face image editing
SF18011016LookinGood: enhancing performance capture with real-time neural re-rendering
SF18011017CariGANs: unpaired photo-to-caricature translation
SF18011018Warp-guided GANs for single-photo facial animation
SF18011021Modeling hair from an RGB-D camera
SF18011022Appearance capture and modeling of human teeth
SF18011023Semantic soft segmentation
SF18011024FaceVR: Real-Time Gaze-Aware Facial Reenactment in Virtual Reality
SF18011025Video to fully automatic 3D hair model
SF18011026Learning to reconstruct shape and spatially-varying reflectance from a single image
SF18011027Invertible grayscale
SF18011028paGAN: real-time avatars using dynamic textures
SF180110293D hair synthesis using volumetric variational autoencoders
SF18011030Two-stage sketch colorization

 

 

SIGGRAPH 2018 papers: 20

https://dl.acm.org/citation.cfm?id=3197517

http://kesen.realtimerendering.com/sig2018.html

  1. Exposure: A White-Box Photo Post-Processing Framework [TOG]
  2. Deep exemplar-based colorization
  3. Deep context-aware descreening and rescreening of halftone images
  4. Non-stationary texture synthesis by adversarial expansion
  5. Synthetic depth-of-field with a single-camera mobile phone SF18011001
  6. Stereo magnification: learning view synthesis using multiplane images SF18011012
  7. FaceVR: Real-Time Gaze-Aware Facial Reenactment in Virtual Reality [TOG] SF18011024
  8. Deep appearance models for face rendering SF18011008
  9. Semantic soft segmentation SF18011023
  10. Faceshop: deep sketch-based face image editing SF18011015
  11. Instant 3D photography SF18011011
  12. Fast and deep deformation approximations
  13. Denoising with kernel prediction and asymmetric loss functions SF18011003
  14. Visemenet: audio-driven animator-centric speech animation
  15. High-fidelity facial reflectance and geometry inference from an unconstrained image
  16. Deep video portraits
  17. Headon: real-time reenactment of human portrait videos SF18011002
  18. Portrait Lighting Transfer Using a Mass Transport Approach [TOG]
  19. MonoPerfCap: Human Performance Capture From Monocular Video [TOG]
  20. Toonsynth: example-based synthesis of hand-colored cartoon animations SF18011004

 

SIGGRAPH Asia 2018 papers: 20

https://dl.acm.org/citation.cfm?id=3272127

http://kesen.realtimerendering.com/siga2018Papers.htm

https://sa2018.siggraph.org/en/attendees/technical-papers

  1. Modeling hair from an RGB-D camera SF18011021
  2. Video to fully automatic 3D hair model SF18011025
  3. Appearance capture and modeling of human teeth SF18011022
  4. 3D hair synthesis using volumetric variational autoencoders SF18011029
  5. Warp-guided GANs for single-photo facial animation SF18011018
  6. Practical dynamic facial appearance modeling and acquisition SF18011006
  7. Stabilized real-time face tracking via a learned dynamic rigidity prior SF18011007
  8. Deep incremental learning for efficient high-fidelity face tracking SF18011010
  9. Deep unsupervised pixelization SF18011005
  10. CariGANs: unpaired photo-to-caricature translation SF18011017
  11. DeepLens: shallow depth of field from a single image
  12. Invertible grayscale SF18011027
  13. LookinGood: enhancing performance capture with real-time neural re-rendering SF18011016
  14. paGAN: real-time avatars using dynamic textures SF18011028
  15. Image smoothing via unsupervised learning SF18011009
  16. Image super-resolution via deterministic-stochastic synthesis and local statistical rectification
  17. Two-stage sketch colorization SF18011030
  18. Efficient palette-based decomposition and recoloring of images via RGBXY-space geometry SF18011014
  19. Learning to reconstruct shape and spatially-varying reflectance from a single image SF18011026
  20. Relighting humans: occlusion-aware inverse rendering for full-body human images