<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[Grey_OnO的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/Grey_OnO</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; Grey_OnO]]></copyright><item><title><![CDATA[Git配置教程: 同时配置github和gitlab的SSH]]></title><link>https://blog.youkuaiyun.com/Grey_OnO/article/details/141755145</link><guid>https://blog.youkuaiyun.com/Grey_OnO/article/details/141755145</guid><author>Grey_OnO</author><pubDate>Sat, 31 Aug 2024 17:43:05 +0800</pubDate><description><![CDATA[Git配置教程: 同时配置github和gitlab的SSH，自动识别并免密push和clone]]></description><category></category></item><item><title><![CDATA[Segment Anything(SAM)的哪些后续方法，又快又好？]]></title><link>https://blog.youkuaiyun.com/Grey_OnO/article/details/136288277</link><guid>https://blog.youkuaiyun.com/Grey_OnO/article/details/136288277</guid><author>Grey_OnO</author><pubDate>Sun, 25 Feb 2024 22:19:14 +0800</pubDate><description><![CDATA[Segment Anything(SAM)的哪些后续方法，又快又好？]]></description><category></category></item><item><title><![CDATA[python借助pybind11调用c++（自定义结构体）]]></title><link>https://blog.youkuaiyun.com/Grey_OnO/article/details/135196123</link><guid>https://blog.youkuaiyun.com/Grey_OnO/article/details/135196123</guid><author>Grey_OnO</author><pubDate>Mon, 25 Dec 2023 11:46:59 +0800</pubDate><description><![CDATA[python借助 pybind11调用c++函数（自定义结构体）]]></description><category></category></item><item><title><![CDATA[【论文阅读】【三维场景特殊点云分割】OpenMask3D：Open-Vocabulary 3D Instance Segmentation]]></title><link>https://blog.youkuaiyun.com/Grey_OnO/article/details/134757803</link><guid>https://blog.youkuaiyun.com/Grey_OnO/article/details/134757803</guid><author>Grey_OnO</author><pubDate>Sun, 03 Dec 2023 21:26:32 +0800</pubDate><description><![CDATA[NeurIPS2023:OpenMask3D-开放词汇三维场景点云实例分割。]]></description><category></category></item><item><title><![CDATA[【论文阅读】【三维场景点云分割】SoftGroup for 3D Instance Segmentation on Point Clouds]]></title><link>https://blog.youkuaiyun.com/Grey_OnO/article/details/132952807</link><guid>https://blog.youkuaiyun.com/Grey_OnO/article/details/132952807</guid><author>Grey_OnO</author><pubDate>Sun, 17 Sep 2023 20:44:40 +0800</pubDate><description><![CDATA[通过进行自下而上的软分类和自上而下的细化来实现三维点云分割：允许每个点与多个类关联，以减轻语义预测错误带来的问题。]]></description><category></category></item><item><title><![CDATA[【论文阅读】【3d场景分割】PointGroup: Dual-Set Point Grouping for 3D Instance Segmentation]]></title><link>https://blog.youkuaiyun.com/Grey_OnO/article/details/132795394</link><guid>https://blog.youkuaiyun.com/Grey_OnO/article/details/132795394</guid><author>Grey_OnO</author><pubDate>Sun, 10 Sep 2023 21:06:27 +0800</pubDate><description><![CDATA[一种自下而上的三维场景点云实例分割方法PointGroup (CVPR2020)。思路清晰易懂]]></description><category></category></item><item><title><![CDATA[【论文阅读】【3D场景点云分割】Mask3D: Mask Transformer for 3D Semantic Instance Segmentation]]></title><link>https://blog.youkuaiyun.com/Grey_OnO/article/details/132642681</link><guid>https://blog.youkuaiyun.com/Grey_OnO/article/details/132642681</guid><author>Grey_OnO</author><pubDate>Sat, 02 Sep 2023 21:21:09 +0800</pubDate><description><![CDATA[第一种真正基于 Transformer 的三维点云实例语义分割方法]]></description><category></category></item><item><title><![CDATA[【论文阅读】【场景物体重建】Total3d：一张照片三维重建你的房间]]></title><link>https://blog.youkuaiyun.com/Grey_OnO/article/details/132515134</link><guid>https://blog.youkuaiyun.com/Grey_OnO/article/details/132515134</guid><author>Grey_OnO</author><pubDate>Sat, 26 Aug 2023 19:45:14 +0800</pubDate><description><![CDATA[Total3D:一张照片三维重建你的房间(CVPR2020 Oral)]]></description><category></category></item><item><title><![CDATA[【论文阅读】【3D场景重建】Panoptic 3D Scene Reconstruction]]></title><link>https://blog.youkuaiyun.com/Grey_OnO/article/details/132381608</link><guid>https://blog.youkuaiyun.com/Grey_OnO/article/details/132381608</guid><author>Grey_OnO</author><pubDate>Sat, 19 Aug 2023 16:55:43 +0800</pubDate><description><![CDATA[全景三维重建，单张RGB图像输入，得到Mesh结果]]></description><category></category></item><item><title><![CDATA[Scannet数据集]]></title><link>https://blog.youkuaiyun.com/Grey_OnO/article/details/125826449</link><guid>https://blog.youkuaiyun.com/Grey_OnO/article/details/125826449</guid><author>Grey_OnO</author><pubDate>Sat, 16 Jul 2022 23:06:15 +0800</pubDate><description><![CDATA[室内三维数据集Scannet]]></description><category></category></item><item><title><![CDATA[基于Dockerfile构建深度学习模型（OpenPCdet）]]></title><link>https://blog.youkuaiyun.com/Grey_OnO/article/details/123334559</link><guid>https://blog.youkuaiyun.com/Grey_OnO/article/details/123334559</guid><author>Grey_OnO</author><pubDate>Mon, 07 Mar 2022 17:20:50 +0800</pubDate><description><![CDATA[文章目录

docker的安装(gpu)

Linux上安装显卡驱动

安装docker

安装NVIDIA Container Toolkit

编写dockerfile

Docker镜像与容器的启动，打包导出

遇到的一些问题：

no module name 'pcdet'

libGL.so.1



以OpenPCdet为例

参考:[Docker] 镜像打包训练好的模型_Emery_learning的博客-优快云博客

docker的安装(gpu)

参考：Docker部署深度学习...]]></description><category></category></item><item><title><![CDATA[【论文阅读】【弱监督-3D目标检测】Weakly Supervised 3D Object Detection from Point Clouds]]></title><link>https://blog.youkuaiyun.com/Grey_OnO/article/details/120344008</link><guid>https://blog.youkuaiyun.com/Grey_OnO/article/details/120344008</guid><author>Grey_OnO</author><pubDate>Fri, 17 Sep 2021 10:37:26 +0800</pubDate><description><![CDATA[前言

ACM MM 2020录用的一篇文章。不使用任何真实的3D框来进行训练。通过采用归一化的点云密度来生成目标候选框。不过性能一般，算是第一个吃螃蟹的人

MIT和微软合作的论文，模型简称为VS3D

论文地址：https://arxiv.org/abs/2007.14366

Code地址：https://github.com/Zengyi-Qin/Weakly-Supervised-3D-Object-Detection



解决的问题及对应贡献

一个无监督的三维对象建议模块(UPM)，它使..]]></description><category></category></item><item><title><![CDATA[【论文阅读】【半监督-3D目标检测】-SESS：Self-Ensembling Semi-Supervised 3D Object Detection]]></title><link>https://blog.youkuaiyun.com/Grey_OnO/article/details/120271939</link><guid>https://blog.youkuaiyun.com/Grey_OnO/article/details/120271939</guid><author>Grey_OnO</author><pubDate>Mon, 13 Sep 2021 17:55:12 +0800</pubDate><description><![CDATA[前言：

《SESS: Self-Ensembling Semi-Supervised 3D Object Detection》

CVPR2020 Oral

新加坡国立大学出品

论文地址：https://openaccess.thecvf.com/content_CVPR_2020/papers/Zhao_SESS_Self-Ensembling_Semi-Supervised_3D_Object_Detection_CVPR_2020_paper.pdf

Code地址：GitHub - Na-Z.]]></description><category></category></item><item><title><![CDATA[【论文阅读】【弱监督-3D目标检测】WS3D:Weakly Supervised 3D Object Detection from Lidar Point Cloud]]></title><link>https://blog.youkuaiyun.com/Grey_OnO/article/details/117718340</link><guid>https://blog.youkuaiyun.com/Grey_OnO/article/details/117718340</guid><author>Grey_OnO</author><pubDate>Tue, 08 Jun 2021 23:51:29 +0800</pubDate><description><![CDATA[前言：
《Weakly Supervised 3D Object Detection from Lidar Point Cloud》（WS3D）
ECCV20录用的一篇文章，后面扩展出一篇TPAMI，我在这里就合并两篇文章一起整理介绍了。
北京理工的一位在读博士。
论文地址：https://arxiv.org/abs/2007.11901
Code地址：https://github.com/hlesmqh/WS3D
TPAMI论文地址：https://ieeexplore.ieee.org/document]]></description><category></category></item><item><title><![CDATA[Ubuntu-torch.cuda.is_available() 的返回值为False]]></title><link>https://blog.youkuaiyun.com/Grey_OnO/article/details/109130335</link><guid>https://blog.youkuaiyun.com/Grey_OnO/article/details/109130335</guid><author>Grey_OnO</author><pubDate>Sat, 17 Oct 2020 11:07:13 +0800</pubDate><description><![CDATA[Ubuntu-torch.cuda.is_available() 的返回值为False
按照流程安装了GPU版本的pytorch，但是输入测试代码：
&gt;&gt;&gt;import torch
&gt;&gt;&gt;torch.cuda.is_available()

结果一直返回结果为False，如果提示是这个则代表GPU版本的pytorch无法正常使用。
结合网上其他的原因，检查了一遍，发现我的cuda和cudnn都安装成功了，版本也应该是匹配的。无奈只能回归到官网去寻找解决方案了
官网提示

]]></description><category></category></item></channel></rss>