CV Code | 计算机视觉开源周报 20190505期

 

今天正式将每周末盘点计算机视觉开源代码的环节,改名为计算机视觉开源周报,并为此编排了期号,希望把这个栏目坚持做下去,方便以后期数多了之后大家参考索引。

 

本周最引人瞩目的开源事件就是Google大脑提出的EfficientNet,这篇论文还登上了Google AI Blog,可见官方也认为是得意之作,CV君在论文刊出当天下午就进行了解读,如果你还不了解,欢迎点击查看:

谷歌大脑提出EfficientNet平衡模型扩展三个维度,取得精度-效率的最大化!

 

另外,52CV曾经报道过的

鲁汶大学提出可端到端学习的车道线检测算法

也开源了:

https://github.com/wvangansbeke/LaneDetection_End2End

欢迎做智能驾驶相关的朋友参考。

 

还有,旷视科技的

CVPR 2019 | 旷视提出超分辨率新方法Meta-SR:单一模型实现任意缩放因子

终于开源了!这是超分辨率领域的重要趋势,欢迎follow~(地址在本文的评论区)

 

 

ICML 2019

卷积网络模型扩展,提高精度,降低计算量,减小模型size

EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks

Mingxing Tan, Quoc V. Le

https://arxiv.org/abs/1905.11946v1

https://github.com/tensorflow/tpu/tree/master/models/official/efficientnet

 

RoNIN: Robust Neural Inertial Navigation in the Wild: Benchmark, Evaluations, and New Methods

Hang Yan, Sachini Herath, Yasutaka Furukawa

https://arxiv.org/abs/1905.12853v1

(将开源,还未公布地址)

 

Learning Navigation Subroutines by Watching Videos

Ashish Kumar, Saurabh Gupta, Jitendra Malik

https://arxiv.org/abs/1905.12612v1

https://ashishkumar1993.github.io/subroutines/

 

IJCAI 2019

视频分类

Hallucinating Optical Flow Features for Video Classification

Yongyi Tang, Lin Ma, Lianqiang Zhou

https://arxiv.org/abs/1905.11799v1

https://github.com/YongyiTang92/MoNet-Features

 

Time Series Workshop of ICML 2019

卫星图像时间序列分析

BreizhCrops: A Satellite Time Series Dataset for Crop Type Identification

Marc Rußwurm, Sébastien Lefèvre, Marco Körner

https://arxiv.org/abs/1905.11893v1

https://github.com/TUM-LMF/BreizhCrops

 

在无标签视频中的自监督目标检测

Toward Self-Supervised Object Detection in Unlabeled Videos

Elad Amrani, Rami Ben-Ari, Tal Hakim, Alex Bronstein

https://arxiv.org/abs/1905.11137v1

(将开源,还未公布地址)

LAW: Learning to Auto Weight

Zhenmao Li, Yichao Wu, Ken Chen, Yudong WU, Shunfeng Zhou, Jiaheng Liu, Junjie Yan

https://arxiv.org/abs/1905.11058v1

(将开源,还未公布地址)

 

三维重建

DISN: Deep Implicit Surface Network for High-quality Single-view 3D Reconstruction

Qiangeng Xu, Weiyue Wang, Duygu Ceylan, Radomir Mech, Ulrich Neumann

https://arxiv.org/abs/1905.10711v1

http://github.com/laughtervv/DISN

 

跨分辨率的人脸识别

Cross-Resolution Face Recognition via Prior-Aided Face Hallucination and Residual Knowledge Distillation

Hanyang Kong, Jian Zhao, Xiaoguang Tu, Junliang Xing, Shengmei Shen, Jiashi Feng

https://arxiv.org/abs/1905.10777v1

(将开源,还未公布地址)

 

Selective Transfer with Reinforced Transfer Network for Partial Domain Adaptation

Zhihong Chen, Chao Chen, Zhaowei Cheng, Ke Fang, Xinyu Jin

https://arxiv.org/abs/1905.10756v1

(将开源,还未公布地址)

 

域适应注意力模型用于非监督的跨域人员重识别

Domain Adaptive Attention Model for Unsupervised Cross-Domain Person Re-Identification

Yangru Huang, Peixi Peng, Yi Jin, Junliang Xing, Congyan Lang, Songhe Feng

https://arxiv.org/abs/1905.10529v1

(将开源,还未公布地址)

 

注意力网络

DIANet: Dense-and-Implicit Attention Network

Zhongzhan Huang, Senwei Liang, Mingfu Liang, Haizhao Yang

https://arxiv.org/abs/1905.10671v1

https://github.com/gbup-group/DIANet

 

高效网络推断

Feature Map Transform Coding for Energy-Efficient CNN Inference

Brian Chmiel, Chaim Baskin, Ron Banner, Evgenii Zheltonozhskii, Yevgeny Yermolin, Alex Karbachevsky, Alex M. Bronstein, Avi Mendelson

https://arxiv.org/abs/1905.10830v1

https://github.com/CompressTeam/TransformCodingInference

 

ICIP 2019

基于注意力网络模型的RGBD语义分割

ACNet: Attention Based Network to Exploit Complementary Features for RGBD Semantic Segmentation

Xinxin Hu, Kailun Yang, Lei Fei, Kaiwei Wang

https://arxiv.org/abs/1905.10089v1

https://github.com/anheidelonghu/ACNet

 

拥挤人群计数

PCC Net: Perspective Crowd Counting via Spatial Convolutional Network

Junyu Gao, Qi Wang, Xuelong Li

https://arxiv.org/abs/1905.10085v1

https://github.com/gjy3035/PCC-Net

 

ICIP 2019

Multi-level Texture Encoding and Representation (MuLTER) based on Deep Neural Networks

Yuting Hu, Zhiling Long, Ghassan AlRegib

https://arxiv.org/abs/1905.09907v1

https://github.com/olivesgatech

转载请注明:《CV Code | 计算机视觉开源周报 20190505期