CV Code | 计算机视觉开源周报20191001期

点击我爱计算机视觉标星，更快获取CVML新技术

---

本图出自OpenVSLAM‍

‍

‍总结了过去一周新出的计算机视觉开源代码。

ICCV 2019 临近，不少论文和相应代码公布，也包括其中的WorkShop的工作。

涵盖的方向包括视觉SLAM、基于标记的SLAM、3D 重建、视线跟踪、植物虫害图像检测识别、人体姿态估计、视频目标分割、语义分割等。

SLAM

OpenVSLAM: A Versatile Visual SLAM Framework

Shinya Sumikura, Mikiya Shibuya, Ken Sakurada

ACM Multimedia 2019 Open Source Software Competition

https://arxiv.org/abs/1910.01122v1

https://github.com/xdspacelab/openvslam

SLAM

TagSLAM: Robust SLAM with Fiducial Markers

Bernd Pfrommer, Kostas Daniilidis

https://arxiv.org/abs/1910.00679v1

https://berndpfrommer.github.io/tagslam_web

3D 重建

Learning Continuous 3D Reconstructions for Geometrically Aware Grasping

Mark Van der Merwe, Qingkai Lu, Balakumar Sundaralingam, Martin Matak, Tucker Hermans

ICRA 2020

https://arxiv.org/abs/1910.00983v1

视线跟踪

RITnet: Real-time Semantic Segmentation of the Eye for Gaze Tracking

Aayush K.Chaudhary, Rakshit Kothari, Manoj Acharya, Shusil Dangi, Nitinraj Nair, Reynold Bailey, Christopher Kanan, Gabriel Diaz, Jeff B. Pelz

ICCV 2019 Workshop OpenEDS Semantic Segmentation Challenge for Eye images

https://arxiv.org/abs/1910.00694v1

https://bitbucket.org/eye-ush/ritnet/

植物虫害图像检测与识别

Research on insect pest image detection and recognition based on bio-inspired methods

Loris Nanni, Gianluca Maguolo, Fabio Pancino

https://arxiv.org/abs/1910.00296v1

https://github.com/LorisNanni/

机器人推动行为数据集

Omnipush: accurate, diverse, real-world dataset of pushing dynamics with RGB-D video

Maria Bauza, Ferran Alet, Yen-Chen Lin, Tomas Lozano-Perez, Leslie P. Kaelbling, Phillip Isola, Alberto Rodriguez

IROS 2019

https://arxiv.org/abs/1910.00618v1

https://web.mit.edu/mcube/omnipush-dataset/

基于胶囊网络的半监督视频目标分割

CapsuleVOS: Semi-Supervised Video Object Segmentation Using Capsule Routing

Kevin Duarte, Yogesh S Rawat, Mubarak Shah

ICCV 2019

https://arxiv.org/abs/1910.00132v1

https://github.com/KevinDuarte/CapsuleVOS

物体 6D 姿态估计

CullNet: Calibrated and Pose Aware Confidence Scores for Object Pose Estimation

Kartik Gupta, Lars Petersson, Richard Hartley

ICCV Workshop on Recovering 6D Object Pose, 2019

https://arxiv.org/abs/1909.13476v1

https://github.com/kartikgupta-at-anu/CullNet

用于语义分割的最大方差损失域适应

Domain Adaptation for Semantic Segmentation with Maximum Squares Loss

Minghao Chen, Hongyang Xue, Deng Cai

ICCV 2019

https://arxiv.org/abs/1909.13589v1

https://github.com/ZJULearning/MaxSquareLoss

单网络全人体的姿态估计方法

Single-Network Whole-Body Pose Estimation

Gines Hidalgo, Yaadhav Raaj, Haroon Idrees, Donglai Xiang, Hanbyul Joo, Tomas Simon, Yaser Sheikh

ICCV 2019

https://arxiv.org/abs/1909.13423v1

https://github.com/CMU-Perceptual-Computing-Lab/openpose_train

OpenPose 升级，CMU提出首个单网络全人体姿态估计网络，速度大幅提高

EdgeCNN：CNN用于边缘计算

EdgeCNN: Convolutional Neural Network Classification Model with small inputs for Edge Computing

Shunzhi Yang, Zheng Gong, Kai Ye, Yungen Wei, Zheng Huang, Zhenhua Huang

https://arxiv.org/abs/1909.13522v1

https://github.com/yangshunzhi1994/EdgeCNN

视频目标分割 | 基于可微分掩膜匹配方法

DMM-Net: Differentiable Mask-Matching Network for Video Object Segmentation

Xiaohui Zeng, Renjie Liao, Li Gu, Yuwen Xiong, Sanja Fidler, Raquel Urtasun

ICCV 2019

https://arxiv.org/abs/1909.12471v1

https://github.com/ZENGXH/DMM_Net

实时多目标跟踪

Towards Real-Time Multi-Object Tracking

Zhongdao Wang, Liang Zheng, Yixuan Liu, Shengjin Wang

https://arxiv.org/abs/1909.12605v1

https://github.com/Zhongdao/Towards-Realtime-MOT

业界首个实时多目标跟踪系统开源

3D 人体姿态与形状重建的学习方法

Learning to Reconstruct 3D Human Pose and Shape via Model-fitting in the Loop

Nikos Kolotouros, Georgios Pavlakos, Michael J. Black, Kostas Daniilidis

ICCV 2019

https://arxiv.org/abs/1909.12828v1

https://seas.upenn.edu/~nkolot/projects/spin

一种可学习的树滤波器，用于结构保持的特征变换，嵌入到语义分割网络中，有效改进了分割精度

Learnable Tree Filter for Structure-preserving Feature Transform

Lin Song, Yanwei Li, Zeming Li, Gang Yu, Hongbin Sun, Jian Sun, Nanning Zheng

NeurIPS-2019

https://arxiv.org/abs/1909.12513v1

https://github.com/StevenGrove/TreeFilter-Torch

---

CV细分方向交流群

52CV已经建立多个CV专业交流群，包括：目标跟踪、目标检测、语义分割、姿态估计、人脸识别检测、医学影像处理、超分辨率、神经架构搜索、GAN、强化学习等，扫码添加CV君拉你入群，如已经为CV君其他账号好友请直接私信，

（请务必注明相关方向，比如：目标检测）

喜欢在QQ交流的童鞋，可以加52CV官方QQ群：805388940。

（不会时时在线，如果没能及时通过验证还请见谅）

---

长按关注我爱计算机视觉