Papers with Code 2020 全年回顾（顶流论文+顶流代码+Benchmarks）

点击上方“CVer”，选择加"星标"置顶

重磅干货，第一时间送达

本文转载自：AI公园

作者：Ross Taylor  编译：ronghuaiyang

导读

2020年Papers with Code 中最顶流的论文，代码和benchmark。

Papers with Code 中收集了各种机器学习的内容：论文，代码，结果，方便发现和比较。通过这些数据，我们可以了解ML社区中，今年哪些东西最有意思。下面我们总结了2020年最热门的带代码的论文、代码库和benchmark。

2020顶流论文

Tan等人的EfficientDet是2020年在Papers with Code上被访问最多的论文。

1. EfficientDet: Scalable and Efficient Object Detection — Tan et al https://paperswithcode.com/paper/efficientdet-scalable-and-efficient-object

2. Fixing the train-test resolution discrepancy — Touvron et al https://paperswithcode.com/paper/fixing-the-train-test-resolution-discrepancy-2

3. ResNeSt: Split-Attention Networks — Zhang et al https://paperswithcode.com/paper/resnest-split-attention-networks

4. Big Transfer (BiT) — Kolesnikov et al https://paperswithcode.com/paper/large-scale-learning-of-general-visual

5. Object-Contextual Representations for Semantic Segmentation — Yuan et al https://paperswithcode.com/paper/object-contextual-representations-for

6. Self-training with Noisy Student improves ImageNet classification — Xie et al https://paperswithcode.com/paper/self-training-with-noisy-student-improves

7. YOLOv4: Optimal Speed and Accuracy of Object Detection — Bochkovskiy et al https://paperswithcode.com/paper/yolov4-optimal-speed-and-accuracy-of-object

8. An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale — Dosovitskiy et al https://paperswithcode.com/paper/an-image-is-worth-16x16-words-transformers-1

9. Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer — Raffel et al https://paperswithcode.com/paper/exploring-the-limits-of-transfer-learning

10. Hierarchical Multi-Scale Attention for Semantic Segmentation — Tao et al https://paperswithcode.com/paper/hierarchical-multi-scale-attention-for

上述很多论文CVer都有分享过解读，详见

• 一文读懂EfficientDet
  

• ResNet最强改进版来了！ResNeSt：Split-Attention Networks
  

• 大神接棒，YOLOv4来了！
  

• 计算机视觉"新"范式: Transformer

2020顶流代码库

Transformers是2020年在Papers with Code上被访问最多的代码库

1. Transformers — Hugging Face — https://github.com/huggingface/transformers

2. PyTorch Image Models — Ross Wightman — https://github.com/rwightman/pytorch-image-models

3. Detectron2 — FAIR — https://github.com/facebookresearch/detectron2

4. InsightFace — DeepInsight — https://github.com/deepinsight/insightface

5. Imgclsmob — osmr — https://github.com/osmr/imgclsmob

6. DarkNet — pjreddie — https://github.com/pjreddie/darknet

7. PyTorchGAN — Erik Linder-Norén — https://github.com/eriklindernoren/PyTorch-GAN

8. MMDetection — OpenMMLab — https://github.com/open-mmlab/mmdetection

9. FairSeq — PyTorch — https://github.com/pytorch/fairseq

10. Gluon CV — DMLC — https://github.com/dmlc/gluon-cv

2020顶流Benchmarks

ImageNet是2020年在Papers with Code上访问最多的benchmark

1. ImageNet — Image Classification — https://paperswithcode.com/sota/image-classification-on-imagenet

2. COCO — Object Detection / Instance Segmentation — https://paperswithcode.com/sota/object-detection-on-coco

3. Cityscapes — Semantic Segmentation — https://paperswithcode.com/sota/semantic-segmentation-on-cityscapes

4. CIFAR-10 — Image Classification — https://paperswithcode.com/sota/image-classification-on-cifar-10

5. CIFAR-100 — Image Classification — https://paperswithcode.com/sota/image-classification-on-cifar-100

6. PASCAL VOC 2012 — Semantic Segmentation — https://paperswithcode.com/sota/semantic-segmentation-on-pascal-voc-2012

7. MPII Human Pose — Pose Estimation — https://paperswithcode.com/sota/pose-estimation-on-mpii-human-pose

8. Market-1501 — Person Re-Identification — https://paperswithcode.com/sota/person-re-identification-on-market-1501

9. MNIST — Image Classification — https://paperswithcode.com/sota/image-classification-on-mnist

10. Human 3.6M — Human Pose Estimation -https://paperswithcode.com/sota/pose-estimation-on-mpii-human-pose

英文原文：https://medium.com/paperswithcode/papers-with-code-2020-review-938146ab9658

CV资源下载

后台回复：CVPR2020，即可下载代码开源的论文合集

后台回复：ECCV2020，即可下载代码开源的论文合集

后台回复：YOLO，即可下载YOLOv4论文和代码

重磅！CVer-细分垂直交流群成立

扫码添加CVer助手，可申请加入CVer-细分垂直方向 微信交流群，也可申请加入CVer大群，细分方向已涵盖：目标检测、图像分割、目标跟踪、人脸检测&识别、OCR、姿态估计、超分辨率、SLAM、医疗影像、Re-ID、GAN、NAS、深度估计、自动驾驶、强化学习、车道线检测、模型剪枝&压缩、去噪、去雾、去雨、风格迁移、遥感图像、行为识别、视频理解、图像融合、图像检索、论文投稿&交流、PyTorch和TensorFlow等群。

一定要备注：研究方向+地点+学校/公司+昵称（如目标检测+上海+上交+卡卡），根据格式备注，才能通过且邀请进群

▲长按加微信群

▲长按关注CVer公众号

整理不易，请给CVer点赞和在看！