cvpr2020 人脸检测与识别_CVPR2020| 最新CVPR2020论文抢先看，附全部下载链接

CVPR 2020

目标检测

Bridging the Gap Between Anchor-based and Anchor-free Detection via Adaptive Training Sample Selection 论文地址：https://arxiv.org/abs/1912.02424

代码：https://github.com/sfzhang15/ATSS

Few-Shot Object Detection with Attention-RPN and Multi-Relation Detector 论文地址：https://arxiv.org/abs/1908.01998

图像分割

Semi-Supervised Semantic Image Segmentation with Self-correcting Networks 论文地址：https://arxiv.org/abs/1811.07073

Deep Snake for Real-Time Instance Segmentation 论文地址：https://arxiv.org/abs/2001.01629

SketchGCN: Semantic Sketch Segmentation with Graph Convolutional Networks 论文地址：https://arxiv.org/abs/2003.00678

PolarMask: Single Shot Instance Segmentation with Polar Representation 论文地址：https://arxiv.org/abs/1909.13226 代码：https://github.com/xieenze/PolarMask

xMUDA: Cross-Modal Unsupervised Domain Adaptation for 3D Semantic Segmentation 论文地址：https://arxiv.org/abs/1911.12676

BlendMask: Top-Down Meets Bottom-Up for Instance Segmentation 论文地址：https://arxiv.org/abs/2001.00309

人脸识别

Towards Universal Representation Learning for Deep Face Recognition 论文地址：https://arxiv.org/abs/2002.11841

Suppressing Uncertainties for Large-Scale Facial Expression Recognition

论文地址：https://arxiv.org/abs/2002.10392 代码：https://github.com/kaiwang960112/Self-Cure-Network

3.Face X-ray for More General Face Forgery Detection 论文地址：https://arxiv.org/pdf/1912.13458.pdf

目标跟踪

1.ROAM: Recurrently Optimizing Tracking Model 论文地址：https://arxiv.org/abs/1907.12006

三维点云&重建

PF-Net: Point Fractal Network for 3D Point Cloud Completion 论文地址：https://arxiv.org/abs/2003.00410

PointAugment: an Auto-Augmentation Framework for Point Cloud Classification 论文地址：https://arxiv.org/abs/2002.10876 代码：https://github.com/liruihui/PointAugment/

3.Learning multiview 3D point cloud registration 论文地址：https://arxiv.org/abs/2001.05119

C-Flow: Conditional Generative Flow Models for Images and 3D Point Clouds 论文地址：https://arxiv.org/abs/1912.07009

RandLA-Net: Efficient Semantic Segmentation of Large-Scale Point Clouds 论文地址：https://arxiv.org/abs/1911.11236

Total3DUnderstanding: Joint Layout, Object Pose and Mesh Reconstruction for Indoor Scenes from a Single Image 论文地址：https://arxiv.org/abs/2002.12212

Implicit Functions in Feature Space for 3D Shape Reconstruction and Completion 论文地址：https://arxiv.org/abs/2003.01456

In Perfect Shape: Certifiably Optimal 3D Shape Reconstruction from 2D Landmarks 论文地址：https://arxiv.org/pdf/1911.11924.pdf

姿态估计

VIBE: Video Inference for Human Body Pose and Shape Estimation 论文地址：https://arxiv.org/abs/1912.05656

代码：https://github.com/mkocabas/VIBE

Distribution-Aware Coordinate Representation for Human Pose Estimation 论文地址：https://arxiv.org/abs/1910.06278

代码：https://github.com/ilovepose/DarkPose

4D Association Graph for Realtime Multi-person Motion Capture Using Multiple Video Cameras 论文地址：https://arxiv.org/abs/2002.12625

Optimal least-squares solution to the hand-eye calibration problem 论文地址：https://arxiv.org/abs/2002.10838

D3VO: Deep Depth, Deep Pose and Deep Uncertainty for Monocular Visual Odometry 论文地址：https://arxiv.org/abs/2003.01060

Multi-Modal Domain Adaptation for Fine-Grained Action Recognition 论文地址：https://arxiv.org/abs/2001.09691

Distribution Aware Coordinate Representation for Human Pose Estimation 论文地址：https://arxiv.org/abs/1910.06278

The Devil is in the Details: Delving into Unbiased Data Processing for Human Pose Estimation 论文地址：https://arxiv.org/abs/1911.07524

9.PVN3D: A Deep Point-wise 3D Keypoints Voting Network for 6DoF Pose Estimation 论文地址：https://arxiv.org/abs/1911.04231

GAN

Your Local GAN: Designing Two Dimensional Local Attention Mechanisms for Generative Models 论文地址：https://arxiv.org/abs/1911.12287 代码：https://github.com/giannisdaras/ylg

MSG-GAN: Multi-Scale Gradient GAN for Stable Image Synthesis 论文地址：https://arxiv.org/abs/1903.06048

Robust Design of Deep Neural Networks against Adversarial Attacks based on Lyapunov Theory 论文地址：https://arxiv.org/abs/1911.04636

小样本&零样本

Improved Few-Shot Visual Classification 论文地址：https://arxiv.org/pdf/1912.03432.pdf

2.Meta-Transfer Learning for Zero-Shot Super-Resolution 论文地址：https://arxiv.org/abs/2002.12213

弱监督&无监督

Rethinking the Route Towards Weakly Supervised Object Localization 论文地址：https://arxiv.org/abs/2002.11359

NestedVAE: Isolating Common Factors via Weak Supervision 论文地址：https://arxiv.org/abs/2002.11576

3.Unsupervised Reinforcement Learning of Transferable Meta-Skills for Embodied Navigation 论文地址：https://arxiv.org/abs/1911.07450

4.Disentangling Physical Dynamics from Unknown Factors for Unsupervised Video Prediction 论文地址：https://arxiv.org/abs/2003.01460

神经网络

Watch your Up-Convolution: CNN Based Generative Deep Neural Networks are Failing to Reproduce Spectral 论文地址：https://arxiv.org/abs/2003.01826

模型加速

GPU-Accelerated Mobile Multi-view Style Transfer 论文地址：https://arxiv.org/abs/2003.00706

视觉常识

What it Thinks is Important is Important: Robustness Transfers through Input Gradients 论文地址：https://arxiv.org/abs/1912.05699

2.Attentive Context Normalization for Robust Permutation-Equivariant Learning 论文地址：https://arxiv.org/abs/1907.02545

Transferring Dense Pose to Proximal Animal Classes 论文地址：https://arxiv.org/abs/2003.00080

Representations, Metrics and Statistics For Shape Analysis of Elastic Graphs 论文地址：https://arxiv.org/abs/2003.00287

Learning in the Frequency Domain 论文地址：https://arxiv.org/abs/2002.12416

7.Filter Grafting for Deep Neural Networks 论文地址：https://arxiv.org/pdf/2001.05868.pdf

8.ClusterFit: Improving Generalization of Visual Representations 论文地址：https://arxiv.org/abs/1912.03330

9.Social-STGCNN: A Social Spatio-Temporal Graph Convolutional Neural Network for Human Trajectory Prediction 论文地址：https://arxiv.org/abs/2002.11927

10.Auto-Encoding Twin-Bottleneck Hashing 论文地址：https://arxiv.org/abs/2002.11930

11.Learning Representations by Predicting Bags of Visual Words 论文地址：https://arxiv.org/abs/2002.12247

12.Holistically-Attracted Wireframe Parsing 论文地址：https://arxiv.org/abs/2003.01663

13.A General and Adaptive Robust Loss Function 论文地址：https://arxiv.org/abs/1701.03077

14.A Characteristic Function Approach to Deep Implicit Generative Modeling 论文地址：https://arxiv.org/abs/1909.07425

15.AdderNet: Do We Really Need Multiplications in Deep Learning? 论文地址：https://arxiv.org/pdf/1912.13200

16.12-in-1: Multi-Task Vision and Language Representation Learning 论文地址：https://arxiv.org/abs/1912.02315

17.Making Better Mistakes: Leveraging Class Hierarchies with Deep Networks 论文地址：https://arxiv.org/abs/1912.09393

18.CARS: Contunuous Evolution for Efficient Neural Architecture Search 论文地址：https://arxiv.org/pdf/1909.04977.pdf 代码：https://github.com/huawei-noah/CARS

19.Towards Learning a Generic Agent for Vision-and-Language Navigation via Pre-training 论文地址：https://arxiv.org/abs/2002.10638 代码：https://github.com/weituo12321/PREVALENT

1.GhostNet: More Features from Cheap Operations(超越Mobilenet v3的架构) 论文链接：https://arxiv.org/pdf/1911.11907arxiv.org 模型(在ARM CPU上的表现惊人)：https://github.com/iamhankai/ghostnetgithub.com

We beat other SOTA lightweight CNNs such as MobileNetV3 and FBNet.

AdderNet: Do We Really Need Multiplications in Deep Learning? (加法神经网络) 在大规模神经网络和数据集上取得了非常好的表现 论文链接：https://arxiv.org/pdf/1912.13200arxiv.org

A Semi-Supervised Assessor of Neural Architectures (神经网络精度预测器 NAS)

Hit-Detector: Hierarchical Trinity Architecture Search for Object Detection (NAS 检测) backbone-neck-head一起搜索， 三位一体

CARS: Contunuous Evolution for Efficient Neural Architecture Search (连续进化的NAS) 高效，具备可微和进化的多重优势，且能输出帕累托前研

On Positive-Unlabeled Classification in GAN (PU+GAN)

Learning multiview 3D point cloud registration(3D点云) 论文链接：arxiv.org/abs/2001.05119

Multi-Modal Domain Adaptation for Fine-Grained Action Recognition(细粒度动作识别) 论文链接：arxiv.org/abs/2001.09691

Action Modifiers:Learning from Adverbs in Instructional Video 论文链接：arxiv.org/abs/1912.06617

PolarMask: Single Shot Instance Segmentation with Polar Representation(实例分割建模) 论文链接：arxiv.org/abs/1909.13226 论文解读：https://zhuanlan.zhihu.com/p/84890413 开源代码：https://github.com/xieenze/PolarMask

Rethinking Performance Estimation in Neural Architecture Search(NAS) 由于block wise neural architecture search中真正消耗时间的是performance estimation部分，本文针对 block wise的NAS找到了最优参数，速度更快，且相关度更高。

Distribution Aware Coordinate Representation for Human Pose Estimation(人体姿态估计) 论文链接：arxiv.org/abs/1910.06278 Github：https://github.com/ilovepose/DarkPose 作者团队主页：https://ilovepose.github.io/coco/

OCR

图像分类

Self-training with Noisy Student improves ImageNet classification 论文地址：https://arxiv.org/abs/1911.04252

Image Matching across Wide Baselines: From Paper to Practice 论文地址：https://arxiv.org/abs/2003.01587

Towards Robust Image Classification Using Sequential Attention Models 论文地址：https://arxiv.org/abs/1912.02184

视频分析

Rethinking Zero-shot Video Classification: End-to-end Training for Realistic Applications 论文地址：https://arxiv.org/abs/2003.01455

代码：https://github.com/bbrattoli/ZeroShotVideoClassification

Say As You Wish: Fine-grained Control of Image Caption Generation with Abstract Scene Graphs 论文地址：https://arxiv.org/abs/2003.00387

Fine-grained Video-Text Retrieval with Hierarchical Graph Reasoning 论文地址：https://arxiv.org/abs/2003.00392

Object Relational Graph with Teacher-Recommended Learning for Video Captioning 论文地址：https://arxiv.org/abs/2002.11566

Zooming Slow-Mo: Fast and Accurate One-Stage Space-Time Video Super-Resolution 论文地址：https://arxiv.org/abs/2002.11616

Blurry Video Frame Interpolation 论文地址：https://arxiv.org/abs/2002.12259

Hierarchical Conditional Relation Networks for Video Question Answering 论文地址：https://arxiv.org/abs/2002.10698

Action Modifiers:Learning from Adverbs in Instructional Video 论文地址：https://arxiv.org/abs/1912.06617

图像处理

Learning to Shade Hand-drawn Sketches 论文地址：https://arxiv.org/abs/2002.11812

2.Single Image Reflection Removal through Cascaded Refinement 论文地址：https://arxiv.org/abs/1911.06634

3.Generalized ODIN: Detecting Out-of-distribution Image without Learning from Out-of-distribution Data 论文地址：https://arxiv.org/abs/2002.11297

更新

视觉常识R-CNN，Visual Commonsense R-CNN

Out-of-distribution图像检测

模糊视频帧插值，Blurry Video Frame Interpolation

元迁移学习零样本超分

3D室内场景理解

6.从有偏训练生成无偏场景图

自动编码双瓶颈哈希

一种用于人类轨迹预测的社会时空图卷积神经网络

面向面向深度人脸识别的通用表示学习

视觉表示泛化性

减弱上下文偏差

可迁移元技能的无监督强化学习

快速准确时空视频超分

对象关系图Teacher推荐学习的视频captioning

弱监督物体定位路由再思考

通过预培训学习视觉和语言导航的通用代理

GhostNet轻量级神经网络

AdderNet:在深度学习中，我们真的需要乘法吗?

CARS:高效神经结构搜索的持续进化

通过协作式的迭代级联微调来移除单图像中的反射

深度神经网络的滤波嫁接

PolarMask：将实例分割统一到FCN

半监督语义图像分割

通过选择性的特征再生来抵御通用攻击

实时的基于细粒度草图的图像检索

用子问题询问VQA模型

从2D范例中学习神经三维纹理空间

NestedVAE:通过薄弱的监督来隔离共同因素

实现多未来轨迹预测

使用序列注意力模型进行稳健的图像分类