2021即将结束了，你今年读了多少论文？

虽然世界仍在从新冠疫情的破坏中复苏，人们无法向从前那样时常线下相聚、共同探讨交流关于学术领域的最新问题，但AI研究也没有停下跃进的步伐。

转眼就是2021年底了，一年就这么就过去了，时光好像被偷走一样。细细数来，你今年读了多少论文？

一名加拿大博主Louis Bouchard以发布时间为顺序，整理出了近40篇2021年不可错过的优秀论文。整体来看，合集中的论文偏重计算机视觉方向。

 

1. DALL·E: Zero-Shot Text-to-Image Generation by OpenAI

论文链接：https://arxiv.org/pdf/2102.12092.pdf

2. VOGUE: Try-On by StyleGAN Interpolation Optimization by Google等

论文链接：https://vogue-try-on.github.io/static_files/resources/VOGUE-virtual-try-on.pdf

3. Taming Transformers for High-Resolution Image Synthesis by 海德堡大学

论文链接：https://compvis.github.io/taming-transformers/

4. Thinking Fast And Slow in AI by IBM等

论文链接：https://arxiv.org/abs/2010.06002、

5. Automatic detection and quantification of floating marine macro-litter in aerial images by 巴塞罗那大学等

论文链接：https://doi.org/10.1016/j.envpol.2021.116490

6. ShaRF: Shape-conditioned Radiance Fields from a Single View  

论文链接：https://arxiv.org/abs/2102.08860

7. Generative Adversarial Transformers by Stanford&Facebook

论文链接：https://arxiv.org/pdf/2103.01209.pdf

8. We Asked Artificial Intelligence to Create Dating Profiles. Would You Swipe Right?

论文链接：https://studyonline.unsw.edu.au/blog/ai-generated-dating-profile

9. Swin Transformer: Hierarchical Vision Transformer using Shifted Windows by 微软亚研

论文链接：https://arxiv.org/abs/2103.14030v2

10. Image GANs meet Differentiable Rendering for Inverse Graphics and Interpretable 3D Neural Rendering by NVidia等

论文链接：https://arxiv.org/pdf/2010.09125.pdf

11. Deep nets: What have they ever done for vision?

论文链接：https://arxiv.org/abs/1805.04025

12. Infinite Nature: Perpetual View Generation of Natural Scenes from a Single Image by Google

论文链接：https://arxiv.org/pdf/2012.09855.pdf

13. Portable, Self-Contained Neuroprosthetic Hand with Deep Learning-Based Finger Control by 明尼苏达大学

论文链接：https://arxiv.org/abs/2103.13452

14. Total Relighting: Learning to Relight Portraits for Background Replacement by Google

论文链接：

https://augmentedperception.github.io/total_relighting/total_relighting_paper.pdf

15. LASR: Learning Articulated Shape Reconstruction from a Monocular Video 

论文链接：

https://openaccess.thecvf.com/content/CVPR2021/papers/Yang_LASR_Learning_Articulated_Shape_Reconstruction_From_a_Monocular_Video_CVPR_2021_paper.pdf

16. Enhancing Photorealism Enhancement

论文链接：http://vladlen.info/papers/EPE.pdf

17. DefakeHop: A Light-Weight High-Performance Deepfake Detector

论文链接：https://arxiv.org/abs/2103.06929

18. High-Resolution Photorealistic Image Translation in Real-Time: A Laplacian Pyramid Translation Network

论文链接：https://arxiv.org/pdf/2105.09188.pdf

19. Barbershop: GAN-based Image Compositing using Segmentation Masks

论文链接：https://arxiv.org/pdf/2106.01505.pdf

20. TextStyleBrush: Transfer of text aesthetics from a single example

论文链接：https://arxiv.org/abs/2106.08385

21. Animating Pictures with Eulerian Motion Fields

论文链接：https://arxiv.org/abs/2011.15128

22. CVPR 2021 Best Paper Award: GIRAFFE - Controllable Image Generation

论文链接：http://www.cvlibs.net/publications/Niemeyer2021CVPR.pdf

23. GitHub Copilot & Codex: Evaluating Large Language Models Trained on Code

论文链接：https://arxiv.org/pdf/2107.03374.pdf

24. Apple: Recognizing People in Photos Through Private On-Device Machine Learning

论文链接：https://machinelearning.apple.com/research/recognizing-people-photos

25. Image Synthesis and Editing with Stochastic Differential Equations by Stanford&CMU

论文链接：https://arxiv.org/pdf/2108.01073.pdf

26. Sketch Your Own GAN by CMU&MIT

论文链接：https://arxiv.org/abs/2108.02774

27. Tesla's Autopilot Explained by Tesla

视频解读：https://youtu.be/DTHqgDqkIRw

28. Styleclip: Text-driven manipulation of StyleGAN imagery by 希伯来大学等

论文链接：https://arxiv.org/abs/2103.17249

29. TimeLens: Event-based Video Frame Interpolation by 华为苏黎世等

论文链接：http://rpg.ifi.uzh.ch/docs/CVPR21_Gehrig.pdf

30. Diverse Generation from a Single Video Made Possible by Weizman

论文链接：https://arxiv.org/abs/2109.08591

31. Skillful Precipitation Nowcasting using Deep Generative Models of Radar by DeepMind

论文链接：https://www.nature.com/articles/s41586-021-03854-z

32. The Cocktail Fork Problem: Three-Stem Audio Separation for Real-World Soundtracks by 三菱

论文链接：https://arxiv.org/pdf/2110.09958.pdf

33. ADOP: Approximate Differentiable One-Pixel Point Rendering

论文链接：https://arxiv.org/pdf/2110.06635.pdf

34. (Style)CLIPDraw: Coupling Content and Style in Text-to-Drawing Synthesis

论文链接：https://arxiv.org/abs/2106.14843

35. SwinIR: Image restoration using swin transformer by ETH

论文链接：https://arxiv.org/abs/2108.10257

36. EditGAN: High-Precision Semantic Image Editing by NVidia等

论文链接：https://arxiv.org/abs/2111.03186

37. CityNeRF: Building NeRF at City Scale by 港中文等

论文链接：https://arxiv.org/pdf/2112.05504.pdf

38. ClipCap: CLIP Prefix for Image Captioning by 特拉维夫大学

论文链接：https://arxiv.org/abs/2111.09734

39. Highly accurate protein structure prediction with AlphaFold

论文链接：https://www.nature.com/articles/s41586-021-03819-2

内容中包含的图片若涉及版权问题，请及时与我们联系删除