ECCV 2020 实例分割+全景分割论文大盘点

网友投稿 635 2022-05-28

前言

计算机视觉Daily 正式系列整理 ECCV 2020的大盘点工作,本文为第三篇:实例分割和全景分割方向。

前两篇详见:

ECCV 2020 目标检测论文大盘点(49篇论文)

ECCV 2020 语义分割论文大盘点(37篇论文)

本文主要包含:实例分割、全景分割等方向。论文PDF已打包好,在后台回复:ECCV2020实例全景分割,即可下载这14篇论文。

▲长按关注我们

注意1:并不包含3D点云分割、视频目标分割,因为这些方向的论文也是超级多的,后续计算机视觉Daily会专门系统整理,还请关注后续内容。

注意2:全景分割的工作并不多

注意3:阿德莱德大学沈春华团队有两篇2D 实例分割工作

目录

实例分割

视频实例分割

弱监督实例分割

其他实例分割

全景分割

实例分割

Conditional Convolutions for Instance Segmentation

论文:https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/1105_ECCV_2020_paper.php

代码:https://github.com/aim-uofa/AdelaiDet

中文解读: ECCV 2020 Oral | 沈春华团队新作CondInst:将条件卷积引入实例分割

Point-Set Anchors for Object Detection, Instance Segmentation and Pose Estimation

论文:https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/1162_ECCV_2020_paper.php

代码:暂无

中文解读: ECCV 2020 | 微软&北大提出Point-set Anchor:统一目标检测,实例分割,以及人体姿态估计

Learning with Noisy Class Labels for Instance Segmentation

论文:https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2062_ECCV_2020_paper.php

代码:https://github.com/longrongyang/LNCIS

中文解读:暂无

LevelSet R-CNN: A Deep Variational Method for Instance Segmentation

论文:https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/4368_ECCV_2020_paper.php

代码:暂无

中文解读:暂无

Supervised Edge Attention Network for Accurate Image Instance Segmentation

论文:https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/5884_ECCV_2020_paper.php

代码:https://github.com//IPIU-detection/SEANet

中文解读:暂无

SOLO: Segmenting Objects by Locations

论文:https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3082_ECCV_2020_paper.php

代码:https://github.com/aim-uofa/AdelaiDet

代码2:https://github.com/WXinlong/SOLO

中文解读(建议结合SOLOv2一起看,SOLOv2应该是收录到NeurIPS 2020了): 更快,更强!SOLOv2来了!实时实例分割新SOTA

视频实例分割

STEm-Seg: Spatio-temporal Embeddings for Instance Segmentation in Videos

论文:https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/1299_ECCV_2020_paper.php

代码:https://github.com/sabarim/STEm-Seg

中文解读:暂无

SipMask: Spatial Information Preservation for Fast Image and Video Instance Segmentation

ECCV 2020 实例分割+全景分割论文大盘点

论文:https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2057_ECCV_2020_paper.php

代码:https://github.com/JialeCao001/SipMask

中文解读:暂无

弱监督实例分割

Commonality-Parsing Network across Shape and Appearance for Partially Supervised Instance Segmentation

论文:https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/596_ECCV_2020_paper.php

代码:https://github.com/fanq15/FewX

中文解读:暂无

论文:https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/6083_ECCV_2020_paper.php

代码:暂无

中文解读:暂无

其他实例分割

The Devil is in Classification: A Simple Framework for Long-tail Instance Segmentation

论文:https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2257_ECCV_2020_paper.php

代码:https://github.com/twangnh/SimCal

中文解读:暂无

PatchPerPix for Instance Segmentation

论文:https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/4968_ECCV_2020_paper.php

代码:https://github.com/Kainmueller-Lab/PatchPerPix

中文解读:暂无

全景分割

Joint Semantic Instance Segmentation on Graphs with the Semantic Mutex Watershed

论文:https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/5393_ECCV_2020_paper.php

代码:暂无

中文解读:暂无

Axial-DeepLab: Stand-Alone Axial-Attention for Panoptic Segmentation

论文:https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/1564_ECCV_2020_paper.php

代码:https://github.com/csrhddlam/axial-deeplab

中文解读:暂无

AI

版权声明:本文内容由网络用户投稿,版权归原作者所有,本站不拥有其著作权,亦不承担相应法律责任。如果您发现本站中有涉嫌抄袭或描述失实的内容,请联系我们jiasou666@gmail.com 处理,核实后本网站将在24小时内删除侵权内容。

上一篇:Android Glide 缓存机制及源码
下一篇:Redis 的持久化方案
相关文章