”兴智杯“全国人工智能创新应用大赛：论文复现赛

[paddle-lwfx]

hi，大家好，非常高兴的告诉大家，兴智杯——前沿模型复现赛已经开始了，本次复现赛共将有15篇的经典&前沿论文供大家复现。详细信息可以参考兴智杯官网链接，或AI Studio 链接，大家是否已经迫不及待了呢~

为了帮助大家更好的了解每篇论文的复现进度，本次复现赛将在本帖中汇总信息。如果你报名/提交结果，需要按格式在进行回复，我们每天会汇总所有的信息，并更新在下表中，你可以通过下面的表格中，了解到每一篇论文的相关信息。

回复的格式为：

【队名】：在兴智杯官网报名时使用的团队名称（请注意一致性，如不一致视为无效作品）

【序号】：你队伍想要复现的论文序号

【论文】：你队伍想要复现的论文名称

【状态】：报名/提交

【repo链接】：repo的链接

如：

【队名】：百度飞桨

【序号】：兴1

【论文】：Paper 1

【状态】：提交

【repo链接】：https://github.com/PaddlePaddle/Paddle

兴智杯 全国人工智能创新应用大赛

兴智杯报名入口

序号	论文信息	复现等级	报名队伍	完成队伍
~~兴1~~	~~Rethinking Bottleneck Structure for Efficient Mobile Network Design~~	⭐️	那拒绝了(报名)(rethinking_bottleneck_design_paddlepaddle) 拼手速(报名&提交)(MobileNeXt-paddle)	拼手速
兴2	Augmented Geometric Distillation for Data-Free Incremental Person ReID	⭐️	SOTAFinder(报名)(Augmented-Geometric-Distillation-PaddlePaddle)just try it(报名)(Augmented-Geometric-Distillation-PaddlePaddle)
兴3	Swin-Unet: Unet-like Pure Transformer for Medical Image Segmentation	⭐️	AIEX(报名)(Swin-Unet-PaddlePaddle) 在校菜鸟(报名)(xingzhibei) 科创生活-吉大小队(报名&提交)(SwinUNet)调查乌度卡(报名&提交)(SwinUNet)
兴4	KiU-Net: Overcomplete Convolutional Architectures for Biomedical Image and Volumetric Segmentation	⭐️	三傻大闹(报名 & 提交)(sanshadanao) SaltyFish(报名)(KiU-Net_PaddlePaddle)
兴5	A Late Fusion CNN for Digital Matting	⭐️	TeamYang(报名)(Late-Fusion-CNN)
~~兴6~~	~~FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flows~~	⭐️	EMA(报名&提交)(fastflow_paddle)	EMA
兴7	Aggregated Contextual Transformations for High-Resolution Image Inpainting	⭐️	FutureSI(报名&提交)(README.md)
兴8	Simple Baselines for Image Restoration	⭐️	low-level(报名)(NAFNet-Paddle.git)不爱做科研(报名&提交)(NAFNet-pd)
兴 9	FCFR-Net: Feature Fusion based Coarse-to-Fine Residual Learning forDepth Completion	⭐️	Inspiration(报名)(Paddle_FCFR_Net)
~~兴10~~	~~CoST: Contrastive Learning of Disentangled Seasonal-Trend Representations for Time Series Forecasting~~	⭐️	aptx1231(报名&提交)(CoST_Paddle)
兴 11	TS2Vec: Towards Universal Representation of Time Series	⭐️	aptx1231(报名)(ts2vec_Paddle)
兴12	Is Pseudo-Lidar needed for Monocular 3D Object detection?	⭐️
兴13	PAConv: Position Adaptive Convolution with Dynamic Kernel Assembling on Point Clouds	⭐️	txyugood(报名 & 提交)(PaddlePAConv)	txyugood
兴14	From Two to One: A New Scene Text Recognizer with Visual Language Modeling Network	⭐️	外阜在京劳动力(报名)(FromTwotToOne)
兴15	Scene Text Telescope: Text-Focused Scene Image Super-Resolution	⭐️	lfy(报名)(scene-text-telescope) 小星(报名)(SmallStar)

[paddle-bot[bot]]

您好，我们已经收到了您的问题，会安排技术人员尽快解答您的问题，请耐心等待。请您再次检查是否提供了清晰的问题描述、复现代码、环境&版本、报错信息等。同时，您也可以通过查看官网API文档、常见问题、历史Issue、AI社区来寻求解答。祝您生活愉快～

Hi! We've received your issue and please be patient to get responded. We will arrange technicians to answer your questions as soon as possible. Please make sure that you have posted enough message to demo your request. You may also check out the API，FAQ，Github Issue and AI community to get the answer.Have a nice day!

[Dufferent]

【队名】：三傻大闹

【序号】：兴4

【论文】：[KiU-Net: Overcomplete Convolutional Architectures for Biomedical Image and Volumetric Segmentation]

【状态】：报名 & 提交

【repo链接】：https://github.com/Dufferent/sanshadanao

[Ifjmww]

【队名】：SaltyFish

【序号】：兴4

【论文】：KiU-Net: Overcomplete Convolutional Architectures for Biomedical Image and Volumetric Segmentation

【状态】：报名

【repo链接】：https://github.com/Ifjmww/KiU-Net_PaddlePaddle

[yongzheJIN]

【队名】小星 【序号】兴15 【论文】Scene Text Telescope: Text-Focused Scene Image Super-Resolution 【状态】报名 【repo链接】：https://github.com/yongzheJIN/SmallStar

[txyugood]

【队名】：txyugood

【序号】：兴13

【论文】：PAConv: Position Adaptive Convolution with Dynamic Kernel Assembling on Point Clouds

【状态】：报名 & 提交

【repo链接】：https://github.com/txyugood/PaddlePAConv 【大赛官网团队名称】：txyugood

[yuanfengliyuanfengli]

【队名】：科创生活-吉大小队 【序号】：兴3 【论文】：Swin-Unet: Unet-like Pure Transformer for Medical Image Segmentation https://arxiv.org/abs/2105.05537 【状态】：报名&提交 【repo链接】：https://github.com/yuanfengliyuanfengli/UCTrans_paddle_try/

[marshall-dteach]

【队名】：调查乌度卡 【序号】：兴3 【论文】：Swin-Unet: Unet-like Pure Transformer for Medical Image Segmentation https://arxiv.org/abs/2105.05537 【状态】：报名&提交 【repo链接】：https://github.com/marshall-dteach/SwinUNet

[kongdebug]

【队名】：不爱做科研

【序号】：兴8

【论文】：Simple Baselines for Image Restoration

【状态】：提交

【repo链接】：https://github.com/kongdebug/NAFNet-pd

[ctkindle]

【队名】：FutureSI

【序号】：兴7

【论文】：Aggregated Contextual Transformations for High-Resolution Image Inpainting

【状态】：提交

【repo链接】：https://github.com/ctkindle/AOT_GAN_Paddle/blob/main/README.md

[Lieberk]

【队名】：超级码立

【序号】：兴14

【论文】：From Two to One: A New Scene Text Recognizer with Visual Language Modeling Network

【状态】：报名 & 提交

【repo链接】：https://github.com/Lieberk/Paddle-VisionLAN

[marshall-dteach]

• 由于兴智杯官网队伍名为：老1.5，因此将调查乌度卡改名为老1.5* 【队名】：老1.5 【序号】：兴3 【论文】：Swin-Unet: Unet-like Pure Transformer for Medical Image Segmentation https://arxiv.org/abs/2105.05537 【状态】：报名&提交 【repo链接】：https://github.com/marshall-dteach/SwinUNet

[Lieberk]

【队名】：超级码立

【序号】：兴15

【论文】：Scene Text Telescope: Text-Focused Scene Image Super-Resolution

【状态】：报名 & 提交

【repo链接】：https://github.com/Lieberk/Paddle-TextSR-STT

[boly0110]

【队名】：boly

【序号】：兴11

【论文】：TS2Vec: Towards Universal Representation of Time Series

【状态】：报名 & 提交

【repo链接】：https://github.com/boly0110/ts2vec_pp.git

[Li-fAngyU]

【队名】：lfy

【序号】：兴15

【论文】：Scene Text Telescope: Text-Focused Scene Image Super-Resolution

【状态】：提交

【repo链接】：https://github.com/Li-fAngyU/scene-text-telescope

[renmada]

【队名】：scv

【序号】：兴5

【论文】：A Late Fusion CNN for Digital Matting

【状态】：提交

【repo链接】：https://github.com/renmada/FusionMatting-paddle

[duyongqi]

【队名】：存在

【序号】：兴1

【论文】：Rethinking Bottleneck Structure for Efficient Mobile Network Design

【状态】：报名&提交

【repo链接】：https://github.com/duyongqi/MobileNeXt

[aptx1231]

【队名】：aptx1231

【序号】：兴10

【论文】：CoST: Contrastive Learning of Disentangled Seasonal-Trend Representations for Time Series Forecasting

【状态】：报名&提交

【repo链接】：https://github.com/aptx1231/CoST_Paddle

[GuoQuanhao]

【队名】：HaoHao

【序号】：兴5

【论文】：A Late Fusion CNN for Digital Matting

【状态】：报名

【repo链接】：https://github.com/GuoQuanhao/FusionMatting-Paddle

[GuoQuanhao]

【队名】：HaoHao

【序号】：兴5

【论文】：A Late Fusion CNN for Digital Matting

【状态】：报名&提交

【repo链接】：https://github.com/GuoQuanhao/FusionMatting-Paddle

[Cloud-Rambler]

【队名】：just try it

【序号】：兴2

【论文】：Augmented Geometric Distillation for Data-Free Incremental Person ReID

【状态】：报名 & 提交

【repo链接】：https://github.com/Cloud-Rambler/Augmented-Geometric-Distillation-PaddlePaddle

[paddle-bot[bot]]

Since you haven't replied for more than a year, we have closed this issue/pr. If the problem is not solved or there is a follow-up one, please reopen it at any time and we will continue to follow up. 由于您超过一年未回复，我们将关闭这个issue/pr。 若问题未解决或有后续问题，请随时重新打开，我们会继续跟进。