AMiner平臺由清華大學計算機系研發，擁有我國徹底自主知識產權。平臺包含了超過2.3億學術論文/專利和1.36億學者的科技圖譜，提供學者評價、專家發現、智能指派、學術地圖等科技情報專業化服務。系統2006年上線，吸引了全球220個國家/地區1000多萬獨立IP訪問，數據下載量230萬次，年度訪問量超過1100萬，成爲學術搜索和社會網絡挖掘研究的重要數據和實驗平臺。ios
AMiner平臺：https://www.aminer.cn/web
NeurIPS，全稱神經信息處理系統大會(Conference and Workshop on Neural Information Processing Systems)， 是機器學習領域的頂級會議,會議固定在每一年的12月舉行 。NeurIPS2020提交數再次創新高，與去年相比增長了38%，共計達到9454篇，總接收1900篇，其中谷歌以202篇傲視羣雄，清華大學64篇。論文接收率20.09%較去年有所降低。網絡
根據AMiner-NeurIPS 2020詞雲圖，小脈發現Reinforcement Learning、Neural Network、Graph Neural Network、Deep Neural Network等都是今年比較火的Topic，無論是投稿量仍是論文接受率都較高。
本文是知乎用戶lam lau整理的NeurIPS2020底層視覺(Low-Level Vision)相關的一些論文。歡迎你們收藏！app
超分辨率
1.論文標題：LAPAR: Linearly-Assembled Pixel-Adaptive Regression Network for Single Image Super-resolution and Beyond機器學習
論文做者：Wenbo Li (Chinese University of Hong Kong) · Kun Zhou (Shenzhen SmartMore Technology Co., Ltd.) · Lu Qi (The Chinese University of Hong Kong) · Nianjuan Jiang (Shenzhen SmartMore Technology Co., Ltd.) · Jiangbo Lu (Shenzhen SmartMore Technology Co., Ltd.) · Jiaya Jia (CUHK)ide
論文連接：https://www.aminer.cn/pub/5f7fdd328de39f08283979ec?conf=neurips2020svg
2.論文標題：Cross-scale Internal Graph Convolution Network for Image Super-Resolution學習
論文做者：Shangchen Zhou (Nanyang Technological University) · Jiawei Zhang (Sensetime Research) · Wangmeng Zuo (Harbin Institute of Technology) · Chen Change Loy (Nanyang Technological University)ui
論文連接：https://www.aminer.cn/pub/5f7fdd328de39f0828397a49?conf=neurips2020.net
3.論文標題：Unfolding the Alternating Optimization for Blind Super Resolution
論文做者：zhengxiong luo (中國科學院自動化所) · Yan Huang (CRIPAC, CASIA) · Shang Li (CASIA) · Liang Wang (NLPR, China) · Tieniu Tan (Chinese Academy of Sciences)
論文連接：https://www.aminer.cn/pub/5f7d959391e011346ad27e60?conf=neurips2020
圖像去模糊
1.論文標題：Deep Wiener Deconvolution: Wiener Meets Deep Learning for Image Deblurring
論文做者：Jiangxin Dong (Max Planck Institute for Informatics) · Stefan Roth (TU Darmstadt) · Bernt Schiele (Max Planck Institute for Informatics)
論文連接：https://www.aminer.cn/pub/5f7fdd328de39f0828397a10?conf=neurips2020
去噪
1.論文標題：Noise2Same: Optimizing A Self-Supervised Bound for Image Denoising
論文做者：Yaochen Xie (Texas A&M University) · Zhengyang Wang (Texas A&M University) · Shuiwang Ji (Texas A&M University)
論文連接：https://www.aminer.cn/pub/5f7fdd328de39f0828397aa0?conf=neurips2020
2.論文標題：Patch2Self: Denoising Diffusion MRI with Self-Supervised Learning
論文做者：Shreyas Fadnavis (Indiana University Bloomington) · Joshua Batson (CZ Biohub) · Eleftherios Garyfallidis (Indiana University)
論文連接：https://www.aminer.cn/pub/5f7fdd328de39f0828397e2d?conf=neurips2020
這篇是醫療圖像的…
圖像恢復
1.論文標題：Neural Sparse Representation for Image Restoration
論文做者：Yuchen Fan (University of Illinois at Urbana-Champaign) · Jiahui Yu (UIUC) · Yiqun Mei (University of Illinois) · Yulun Zhang (Northeastern University) · Yun Fu (Northeastern University) · Ding Liu (Bytedance AI Lab) · Thomas Huang (University of Illinois)
論文連接：https://www.aminer.cn/pub/5edf5dd891e011bc656dee3c?conf=neurips2020
這篇題目頗有意思啊，二做仍是作inpainting的yujiahui大佬…
2.論文標題：CLEARER: Multi-Scale Neural Architecture Search for Image Restoration
論文做者：Yuanbiao Gou (College of Computer Science, Sichuan University) · Boyun Li (College of Computer Science, Sichuan University) · Zitao Liu (TAL AI Lab) · Songfan Yang (TAL AI Lab) · Xi Peng (Institute for Infocomm, Research Agency for Science, Technology and Research (A*STAR) Singapore)
論文連接：https://www.aminer.cn/pub/5f7fdd328de39f0828397cc6?conf=neurips2020
圖像編輯
1.論文標題：Lightweight Generative Adversarial Networks for Text-Guided Image Manipulation
論文做者：Bowen Li (University of Oxford) · Xiaojuan Qi (The University of Hong Kong) · Philip Torr (University of Oxford) · Thomas Lukasiewicz (University of Oxford)
論文連接：https://www.aminer.cn/pub/5f7fdd328de39f08283979ef?conf=neurips2020
2.論文標題：Swapping Autoencoder for Deep Image Manipulation
論文做者：Taesung Park (UC Berkeley) · Jun-Yan Zhu (Adobe, CMU) · Oliver Wang (Adobe Research) · Jingwan Lu (Adobe Research) · Eli Shechtman (Adobe Research, US) · Alexei Efros (UC Berkeley) · Richard Zhang (Adobe)
論文連接：https://www.aminer.cn/pub/5efdbed591e01191d3d283ef?conf=neurips2020
圖像生成/GAN
1.論文標題：Few-shot Image Generation via Self-Adaptation
論文做者：Yijun Li (Adobe Research) · Richard Zhang (Adobe) · Jingwan (Cynthia) Lu (Adobe Research) · Eli Shechtman (Adobe Research, US)
論文連接：https://www.aminer.cn/pub/5f7fdd328de39f082839799c?conf=neurips2020
2.論文標題：Denoising Diffusion Probabilistic Models
論文做者：Jonathan Ho (UC Berkeley) · Ajay Jain (UC Berkeley) · Pieter Abbeel (UC Berkeley & covariant.ai)
論文連接：https://www.aminer.cn/pub/5ef0816891e0112aee042b88?conf=neurips2020
3.論文標題：DeepI2I: Enabling Deep Hierarchical Image-to-Image Translation by Transferring from GANs
論文做者：yaxing wang (Centre de Visió per Computador (CVC)) · Lu Yu (computer vision center, UAB) · Joost van de Weijer (Computer Vision Center Barcelona)
論文連接：https://www.aminer.cn/pub/5f7fdd328de39f0828397b7c?conf=neurips2020
HDR
1.論文標題：UnModNet: Learning to Unwrap a Modulo Image for High Dynamic Range Imaging
論文做者：Chu Zhou (Peking University) · Hang Zhao (MIT) · Jin Han (Peking University) · Chang Xu (University of Sydney) · Chao Xu (Peking University) · Tiejun Huang (Peking University) · Boxin Shi (Peking University)
論文連接：https://www.aminer.cn/pub/5f7fdd328de39f0828397b02/?conf=neurips2020
文章來源：https://zhuanlan.zhihu.com/p/265979104
本文同步分享在 博客「AMiner科技」（CSDN）。
若有侵權，請聯繫 support@oschina.cn 刪除。
本文參與「OSC源創計劃」，歡迎正在閱讀的你也加入，一塊兒分享。