7

3D视觉关键字获取资料汇总

 2 years ago
source link: https://zhuanlan.zhihu.com/p/514819888
Go to the source link to view the article. You can view the picture content, updated content and better typesetting reading experience. If the link is broken, please click the button below to view the snapshot at that time.
neoserver,ios ssh client

「3D视觉工坊」及「计算机视觉工坊」后台关键字回复汇总

  • 回复关键字:Panop
标题:ICRA2022:自适应环境的移动机器人自主导航 Title:An Efficient Locally Reactive Controller for Safe Navigation in Visual Teach and Repeat Missions 链接:https://arxiv.org/abs/2201.03938
  • 回复关键字:EvIntSR-Net
标题:北大工作!EvIntSR-Net:基于事件相机的视频帧重建和超分辨率(ICCV2021) title:ICCV 2021: EvIntSR-Net: Event Guided Multiple Latent Frames Reconstruction and Super-resolution 链接:https://openaccess.thecvf.com/content/ICCV2021/papers/Han_EvIntSR-Net_Event_Guided_Multiple_Latent_Frames_Reconstruction_and_Super-Resolution_ICCV_2021_paper.pdf
  • 回复关键字:诺亚方舟
简历直通车:[email protected] 邮件主题及附件命名为:姓名+学校+年级+实习开始时间+可实习时间长度
  • 回复关键字:2021SLAM
[资料 | 2021年开源SLAM算法集锦] 链接:https://www.jianguoyun.com/p/DWj0LqsQnL_KCRitnqYE
  • 回复关键字:LMSCNet
标题:LMSCNet:轻量级多尺度 3D 语义分割 Title:LMSCNet: Lightweight Multiscale 3D Semantic Completion 链接:https://arxiv.org/abs/2008.10559 代码:https://github.com/cv-rits/LMSCNet
  • 回复关键字:小目标检测
tph-yolov5:https://github.com/cv516Buaa/tph-yolov5/blob/main/models/common.py 其他改进:https://github.com/Gumpest/YOLOv5-Multibackbone-Compression 增加小目标检测层:https://mp.weixin.qq.com/s/3ecefgmOZU-zugMdMWlTaA
  • 回复关键字:MHPE
标题:麻省理工:解决物体级SLAM中姿态歧义的多假设方法(IROS 2021) Title:A Multi-Hypothesis Approach to Pose Ambiguity in Object-Based SLAM | IROS 2021 链接:http://people.csail.mit.edu/yuewang/publication/obj-slam/
  • 回复关键字:TUPPer
标题:[IROS2021] TUPPer-Map:用于 3D 度量语义建图的时间和统一全景感知 Title:TUPPer-Map: Temporal and Unified Panoptic Perception for 3D Metric-Semantic Mapping 链接:https://ieeexplore.ieee.org/abstract/document/9636599
  • 回复关键字:ESMOT
标题:[ICCV2021] 轻舟智航,探索用于自动驾驶的3D点云多目标跟踪 Title:Exploring Simple 3D Multi-Object Tracking for Autonomous Driving 链接:https://arxiv.org/abs/2108.10312v1
  • 回复关键字:Panop
标题:NeurIPS 2021:利用神经网络从单个 RGB 图像重建全景 3D 场景 Title:Panoptic 3D Scene Reconstruction From a Single RGB Image (NeurIPS 2021) 链接:https://arxiv.org/pdf/2111.02444.pdf
  • 回复关键字:TMWS
标题:[ICCV 2021] 基于水体流动的可微框架进行水下单目3D重建 Title:In-the-Wild Single Camera 3D Reconstruction Through Moving Water Surfaces [ICCV 2021 Oral] 链接:https://vccimaging.org/Publications/Xiong2021MovingWater/Xiong2021MovingWater.pdf 代码:https://github.com/vccimaging/Reconstrution_Through_Moving_Water
  • 回复关键字:无人机
直播主题: 四旋翼飞行器:算法与实战 直播地址: https://app0s6nfqrg6303.h5.xiaoeknow.com/v2/course/alive/l_61bdc052e4b0ed68333fb214?app_id=app0S6nfQRG6303&alive_mode=0&pro_id=&type=2
  • 回复关键字:3D线激光
直播链接: https://app0s6nfqrg6303.h5.xiaoeknow.com/v2/course/alive/l_61b35deae4b0dabb65a398dc?app_id=app0S6nfQRG6303&alive_mode=0&pro_id=&type=2 课件:链接:https://pan.baidu.com/s/1yZPBj6Jr-ZnjUUEPdMt3ow 提取码:w7wx 加入「3D视觉从入门到精通」知识星球,获取更多干货课件及视频。
  • 回复关键字:DeepPanoContext
论文地址:https://arxiv.org/abs/2108.10743v1 项目地址:https://chengzhag.github.io/publication/dpc
  • 回复关键字:ORBSLAM
ORB-SLAM2 / ORB-SLAM3 相关改进代码汇总 PDF链接: https://pan.baidu.com/s/1_rAypo0dBWivxWl61-pgig 提取码: 8t64
  • 回复关键字:ASBM
标题:ICCV2021:基于分配空间的多目标跟踪和分割 Title:Assignment-Space-Based Multi-Object Tracking and Segmentation (ICCV 2021) 代码:https://anwesachoudhuri.github.io/Assignment-Space-based-MOTS/
  • 回复关键字:RSVHSW
标题:[ICCV2021] 单目实时3D人体姿态估计,效果巨好!仿真数据集训练,真实数据测试! Title:2021 Real-time single view 3D human shape and pose with uncertainty in the wild 链接:https://arxiv.org/abs/2110.00990 代码:https://github.com/akashsengupta1997/HierarchicalProbabilistic3DHuman
  • 回复关键字:VisualE
标题:[ECCV2020] Facebook研究,AI实现生物声呐定位、导航 Title:VisualEchoes: Spatial Image Representation Learning through Echolocation 链接:https://arxiv.org/pdf/2005.01616.pdf 项目:https://vision.cs.utexas.edu/projects/visualEchoes/
  • 回复关键字:CVPR-AEC
Title:香港大学工作!CVPR2022:事件相机的自动对焦(代码开源) 标题:Autofocus for Event Cameras 链接:https://arxiv.org/pdf/2203.12321.pdf 代码:https://eleboss.github.io/eaf_webpage/
  • 回复关键字:Occlusion
Title:OcclusionFusion: Occlusion-aware Motion Estimation for Real-time Dynamic 3D Reconstruction 标题:OcclusionFusion:遮挡感知运动估计的实时动态3D重建 链接:https://arxiv.org/abs/2203.07977 代码:https://wenbin-lin.github.io/OcclusionFusion/
  • 回复关键字:DHDRU
标题:[CVPR2022] 使用单幅彩色点投影的深度学习高光谱和深度重建 Title:Deep Hyperspectral-Depth Reconstruction Using Single Color-Dot Projection 链接:http://www.ok.sc.e.titech.ac.jp/res/DHD/ 代码:https://github.com/GrayMask/DHD
  • 回复关键字:诺亚方舟实验室
“2022年实习+姓名”及个人简历发送至邮箱:[email protected]
  • 回复关键字:GCTC
标题:ICRA2022:全局一致且紧密耦合的 3D LiDAR 惯性建图 Title:Globally Consistent and Tightly Coupled 3D LiDAR Inertial Mapping 链接:https://staff.aist.go.jp/k.koide/assets/pdf/icra2022.pdf
  • 回复关键字:GANgeal
标题:CVPR2022:GAN监督的密集视觉配准用于多种AR应用 Title:GANgealing: GAN-Supervised Dense Visual Alignment 链接:https://arxiv.org/abs/2112.05143 代码:https://github.com/wpeebles/gangealing
  • 回复关键字:ICGFR
标题:[CVPR2022] ICG:融合区域和深度以实现无纹理对象的高效3D跟踪 Title:Iterative Corresponding Geometry: Fusing Region and Depth for Highly Efficient 3D Tracking of Textureless Objects 链接:https://arxiv.org/abs/2203.05334 代码:https://github.com/DLR-RM/3DObjectTracking
  • 回复关键字:InsetGAN
标题:[CVPR2022] 能换全身发型、服装的Gan网络(代码开源) Title:InsetGAN for Full-Body Image Generation 代码:https://afruehstueck.github.io/insetgan/
  • 回复关键字:PCICT
标题:CVPR2022:最强鉴别器!用深度学习鉴定伪造的人脸 Title:Protecting Celebrities with Identity Consistency Transformer-CVPR2022 链接:https://arxiv.org/abs/2203.01318
  • 回复关键字:HSC4D
标题:[CVPR2022] 厦门大学空间感知与计算实验室:使用可穿戴 IMU和LiDAR的大规模室内外4D场景捕获 [代码开源] Title:Human-centered 4D Scene Capture in Large-scale Indoor-outdoor Space Using Wearable IMUs and LiDAR 链接:https://arxiv.org/abs/2203.09215 代码:https://github.com/climbingdaily/HSC4D
  • 回复关键字:RefVSR
标题:CVPR2022:广角、超广角和长焦镜头的多摄像头超分辨率合成,1080P变8K视频! Title:Reference-based Video Super-Resolution Using Multi-Camera Video Triplets 链接:https://arxiv.org/abs/2203.14537v1
  • 回复关键字:SAR-Net
标题:复旦大学工作!SAR-Net:用于类别级 6D 对象姿势和大小估计的形状配准和恢复网络 Title:SAR-Net: Shape Alignment and Recovery Network for Category-level 6D Object Pose and Size Estimation 代码:https://hetolin.github.io/SAR-Net/
  • 回复关键字:StyleMesh
标题:慕尼黑工业大学:StyleMesh:室内 3D 场景重建的风格转移 [CVPR2022] Title:StyleMesh: Style Transfer for Indoor 3D Scene Reconstructions 链接:https://arxiv.org/pdf/2112.01530.pdf 代码:https://lukashoel.github.io/stylemesh/
  • 回复关键字:CVPR2022-BOBL
标题:香港城市大学与微软联合工作!AI还原老电影 [CVPR2022] Title:Bringing Old Films Back to Life 链接:https://arxiv.org/pdf/2203.17276.pdf 代码:http://raywzy.com/Old_Film/
  • 回复关键字:CVPR-CRF
标题:用于单目深度估计的神经窗口全连接CRF [CVPR2022] Title:NeW CRFs: Neural Window Fully-connected CRFs for Monocular Depth Estimation 链接:https://arxiv.org/abs/2203.01502 代码:https://weihaosky.github.io/newcrfs/
  • 回复关键字:ELAVM
标题:港大火星实验室:在线 LiDAR 里程计的高效概率自适应体素映射 Title:Efficient and Probabilistic Adaptive Voxel Mapping for Accurate Online LiDAR Odometry 链接:https://arxiv.org/abs/2109.07082v4
  • 回复关键字:数据集
免费数据集地址:http://m6z.cn/5ZBLEU
  • 回复关键字:3DV-ESL
标题:苏黎世联邦理工:基于事件相机的三维结构光技术ESL,性能好于Realsense技术! Title:ESL: Event-based Structured Light 链接:https://rpg.ifi.uzh.ch/docs/3DV21_Muglikar_ESL.pdf 代码:https://github.com/uzh-rpg/ESL
  • 回复关键字:视觉和惯性传感器
论文链接:https://pan.baidu.com/s/1AsYnlQXAFGextg9f78C75g 提取码:f8yk 相关视频和代码:https://Smilels.github.io/multimodal-translation-teleop
  • 回复关键字:CasMVSNet
论文代码:https://github.com/alibaba/cascade-stereo
  • 回复关键字:ACCV2020
链接: https://pan.baidu.com/s/1i6CBoxK6Ar4pHQKFO4ornA 提取码: 5hen
  • 回复关键字:半直接单目
【代码地址】https://github.com/sunghoon031/LCSD_SLAM
  • 回复关键字:RGB检测
论文下载:https://openaccess.thecvf.com/content/ACCV2020/papers/Yan_SDCNet_Size_Divide_and_Conquer_Network_for_Salient_Object_Detection_ACCV_2020_paper.pdf 论文代码:https://github.com/Sssssbo/SDCNet (等待上传)
  • 回复关键字:P-MVSNet
论文链接: https://openaccess.thecvf.com/content_ICCV_2019/papers/Luo_P-MVSNet_Learning_Patch-Wise_Matching_Confidence_Aggregation_for_Multi-View_Stereo_ICCV_2019_paper.pdf
  • 回复关键字:Monodepth2
论文链接:https://sci-hub.se/10.1109/iccv.2019.00393
  • 回复关键字:Fast-MVSNet
开源代码:https://github.com/svip-lab/FastMVSNet
  • 回复关键字:调研报告
《2020年度中国计算机视觉人才调研报告》 链接:https://pan.baidu.com/s/1avr-Rm5aTWX800rVZdF9Ig 提取码:ytg1
  • 回复关键字:U-CMR
项目链接:https://shubham-goel.github.io/ucmr/ 代码链接:https://github.com/shubham-goel/ucmr/tree/master
  • 回复关键字:目标检测竞赛
大盘点 | 2020年5篇目标检测算法最佳综述链接: https://pan.baidu.com/s/1IRZeDN4P8TjWu2MJIc261A 密码:3f90
  • 回复关键字:Camera答案
参考答案地址: https://github.com/ethan-li-coding/StereoV3DCode
  • 回复关键字:坐标系与相机参数
随课代码地址: StereoV3DCode:https://github.com/ethan-li-coding/StereoV3DCode
  • 回复关键字:RGB图像重建
项目地址:https://smalr.is.tue.mpg.de/ 论文地址:http://files.is.tue.mpg.de/black/papers/zuffiCVPR2018.pdf
  • 回复关键字:RepVGG
代码地址: https://github.com/DingXiaoH/RepVGG
  • 回复关键字:3D鸟类重建
代码链接:https://github.com/marcbadger/avian-mesh 项目链接:https://marcbadger.github.io/avian-mesh 论文链接:https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123630001.pdf
  • 回复关键字:HPE
https://arxiv.org/pdf/2006.01423.pdf
  • 回复关键字:BlendedMVS
数据集和预训练模型:https://github.com/YoYo000/BlendedMVS
  • 回复关键字:Bundle Adjustment
本文所有例子代码地址: https://github.com/qxiaofan/awesome-BA-exercise
  • 回复关键字:EF-Net
论文下载:https://www.sciencedirect.com/science/article/abs/pii/S0031320320305434
  • 回复关键字:Speech2Face
项目地址:https://speech2face.github.io/
  • 回复关键字:人脸
链接:https://pan.baidu.com/s/1Kg1CLmB9Aju8ncyg9MYRfQ 密码:mvq1
  • 回复关键字:写作
链接:https://pan.baidu.com/s/1zlc7fBqWfA15kyNCfk5-OQ 提取码:0101
  • 回复关键字:地图管理
论文打包下载: https://pan.baidu.com/s/1kRX7s5GIzOjSAy1eAhi2HQ 提取码: s5ak
  • 回复关键字:综述
CV 综述链接: https://pan.baidu.com/s/1xzJWAVuP8ReceCG27mrKIg 提取码:52CV
  • 回复关键字:尺度可观测性
论文地址:https://arxiv.org/pdf/2103.15215.pdf 代码地址:https://github.com/jpl-x
  • 回复关键字:外参标定
代码即将在https://github.com/alibaba开源
  • 回复关键字:DELTAS
数据集和预训练模型:https://github.com/magicleap/DELTAS
  • 回复关键字:D4LCN
原文地址:https://arxiv.org/pdf/1912.04799v1 官方代码:https://github.com/dingmyu/D4LCN
  • 回复关键字:transformer
Visual Transformer综述下载 链接:https://pan.baidu.com/s/1mfWTjjYnbRsIurORv0MQwg 提取码:favd
  • 回复关键字:三角化
直播地址: https://app0s6nfqrg6303.h5.xiaoeknow.com/v2/course/alive/l_60f0419be4b08f7ad23d3b53?app_id=app0S6nfQRG6303&alive_mode=0&pro_id=&type=2 直播时间:7月25日 周日晚上20:00 直播主题:SLAM \SfM中的两视图三角化算法
  • 回复关键字:深度估计
直播时间:7月17日晚八点 直播地址:https://app0s6nfqrg6303.h5.xiaoeknow.com/v2/course/alive/l_60eeeda8e4b08f7ad23cd33f?app_id=app0S6nfQRG6303&alive_mode=0&pro_id=p_60acfdd9e4b07e4d7fd8bcaf&type=2
  • 回复关键字:高反光
直播时间:7.10周六晚八点 直播地址:https://app0s6nfqrg6303.h5.xiaoeknow.com/v2/course/alive/l_60e5bca5e4b0bd401d89494f?app_id=app0S6nfQRG6303&alive_mode=0&pro_id=p_609281f5e4b0d4eb039721a1&type=2
  • 回复关键字:点云补全
直播地址:https://app0s6nfqrg6303.h5.xiaoeknow.com/v2/course/alive/l_60d9d38ae4b057a4e728db27?app_id=app0S6nfQRG6303&alive_mode=0&pro_id=&type=2 直播时间:7月3日 周六晚20:00
  • 回复关键字:逆相机法
链接:https://pan.baidu.com/s/1SaYMPHj3Mh_A9mQIGx-R8g 提取码:803z
  • 回复关键字:单目自监督
代码地址:https://github.com/nvlabs/umr 论文地址:https://arxiv.org/pdf/2003.06473.pdf
  • 回复关键字:双目结构光
双目结构光,直播地址:https://app0S6nfQRG6303.h5.xiaoeknow.com/v2/course/alive/l_60aface1e4b0c7264219b522?app_id=app0S6nfQRG6303&alive_mode=0&pro_id=p_609281f5e4b0d4eb039721a1&type=2 直播时间:2021年5月30日上午10点
  • 回复关键字:LiTAMIN2
视频链接:https://www.bilibili.com/video/BV1yp4y1h792/ 论文链接: https://arxiv.org/pdf/2103.00784v1.pdf
  • 回复关键字:DRO
论文链接:https://arxiv.org/abs/2103.13201 代码链接:https://github.com/aliyun/dro-sfm
  • 回复关键字:导航系统
论文地址: https://pan.baidu.com/s/1YutBKtSNDNe8FN8BX8Untw 提取码:wc0k 视频地址:https://mp.weixin.qq.com/s/9dPmQ6pd8-vRg8NFv_nNPA
  • 回复关键字:PREDATOR
论文代码:https://github.com/ShengyuH/OverlapPredator
  • 回复关键字:LiDAR里程计
论文地址: https://pan.baidu.com/s/1SZgS6koRGmUsBlkpUneD8w 提取码:v2nl 代码地址:https://github.com/PRBonn/puma 视频地址:https://mp.weixin.qq.com/s/xrFAoBjC8llMcAFeD6PGMA
  • 回复关键字:目标检测综述
综述:目标检测二十年(2001-2021) 链接: https://pan.baidu.com/s/1YJ7x68H7VldF4CDF7HQ7fw 提取码: jmz2
  • 回复关键字:二维图像GAN
原文链接:https://openreview.net/pdf?id=FGqiDsBUKL0 代码地址:https://github.com/XingangPan/GAN2Shape
  • 回复关键字:PHOSA
项目网址:https://jasonyzhang.com/phosa/ 论文网址:https://arxiv.org/pdf/2007.15649.pdf
  • 回复关键字:PAConv
论文链接:https://arxiv.org/abs/2103.14635 代码链接:https://github.com/CVMI-Lab/PAConv
  • 回复关键字:相机定位
论文地址:https://arxiv.org/pdf/2103.16792.pdf Code开源:https://github.com/Tangshitao/DenseScene-Matching
  • 回复关键字:MVSNet网络
开源代码(tensorflow):https://github.com/YoYo000/MVSNet 开源代码(PyTorch):https://github.com/XYZ-qiyh/MVSNet_pytorch
  • 回复关键字:pixelNeRF
项目地址:https://alexyu.net/pixelnerf/ 论文地址:https://arxiv.org/pdf/2012.02190.pdf
  • 回复关键字:LVI-SAM
代码开源:http://github.com/TixiaoShan/LVI-SAM
  • 回复关键字:鲁棒点云
论文代码:https://github.com/fukexue/RGM.
  • 回复关键字:SpinNet
论文代码:https://github.com/QingyongHu/SpinNet
  • 回复关键字:思无界科技
简历请发送到:[email protected] 备注:3D视觉工坊推荐
  • 回复关键字:CenterPoint
代码:https://github.com/tianweiy/CenterPoint 论文:https://arxiv.org/pdf/2006.11275.pdf
  • 回复关键字:PMS
Windows版本:https://github.com/ethan-li-coding/PatchMatchStereo Linux版本:https://github.com/Haonan-DONG/PatchMatchStereo_linux
  • 回复关键字:PISE
项目主页:http://cic.tju.edu.cn/faculty/likun/projects/PISE 代码:https://github.com/Zhangjinso/PISE
  • 回复关键字:单目重建
Mobile3DRecon: Real-time Monocular 3D Reconstruction on a Mobile Phone 链接: https://pan.baidu.com/s/1kaPCPR_eLDEWgoZlWyeuQQ 提取码: d7n4
  • 回复关键字:CorrNet3D
论文链接:https://arxiv.org/pdf/2012.15638.pdf 代码链接:https://github.com/ZENGYIMING-EAMON/CorrNet3D
  • 回复关键字:0009
论文分享(语义分割相关论文) 链接:https://pan.baidu.com/s/1R4pZNK_AkPjaCH2gUuE-tA 提取码:lqka
  • 回复关键字:R3LIVE
论文地址: https://github.com/hku-mars/r3live/blob/master/papers/R3LIVE:%20A%20Robust%2C%20Real-time%2C%20RGB-colored%2C%20LiDAR-Inertial-Visual%20tightly-coupled%20stateEstimation%20and%20mapping%20package.pdf 代码:https://github.com/hku-mars/r3live
  • 回复关键字:车端激光
代码:http://wiki.ros.org/velo2cam_calibration
  • 回复关键字:VIO定位
代码开源:https://github.com/Zumbalamambo/2D-3D-pose-tracking
  • 回复关键字:Range Image
代码开源:https://github.com/PRBonn/range-mcl
  • 回复关键字:SegICP
Code: https://github.com/jaymwong/seline
  • 回复关键字:0010
链接:https://pan.baidu.com/s/18TruEsScJDHY8Cw3S2hRfA 提取码:oir1
  • 回复关键字:数据集
免费数据集地址:http://m6z.cn/5ZBLEU
  • 回复关键字:谷歌X实习生
补充材料视频 https://share.weiyun.com/XbuFfAvcIROS 开会讲解视频 https://share.weiyun.com/R3myU8Rc 讲解视频字幕 https://share.weiyun.com/8bgNvu0U
  • 回复关键字:MVTN
论文地址:https://arxiv.org/abs/2011.13244 项目地址:https://github.com/ajhamdi/MVTN
  • 回复关键字:结构光方法
直播地址: https://app0s6nfqrg6303.h5.xiaoeknow.com/v2/course/alive/l_61166783e4b0bf643005a813?app_id=app0S6nfQRG6303&alive_mode=0&pro_id=p_609281f5e4b0d4eb039721a1&type=2 直播时间:2021年8月15日上午10点
  • 回复关键字:梅卡曼德
联系方式:[email protected] 或者 加微信:CV_LAB, 备注:3D视觉工坊 + 应聘公司和岗位名称
  • 回复关键字:EADM
标题:ISMAR 2020,宜家+苹果,实现混合现实家居设计! Title:ISMAR 2020:Exploiting ARKit Depth Maps for Mixed Reality Home Design 链接:https://drive.google.com/file/d/19cg1hi_odSSnafmNw3NZoKzab_6VY-K0/view
  • 回复关键字:EAOSLA
标题:EAO-SLA:集成数据关联的单目半稠密物体级SLAM,代码开源(IROS2020) Title:EAO-SLAM: Monocular Semi-Dense Object SLAM Based on Ensemble Data Association 文:https://arxiv.org/abs/2004.12730v2 代码:https://github.com/yanmin-wu/EAO-SLAM
  • 回复关键字:REAL
标题:IROS2021,REAL:放出一台无人机对未知路况进行主动闭环的快速探索 Title:REAL: Rapid Exploration with Active Loop-Closing toward Large-Scale 3D Mapping using UAVs 论文:https://arxiv.org/abs/2108.02590
  • 回复关键字:尺度可观测性
论文地址:https://arxiv.org/pdf/2103.15215.pdf 代码地址:https://github.com/jpl-x
  • 回复关键字:外参标定
代码即将在https://github.com/alibaba开源
  • 回复关键字:THFM
标题:最新ICCV2021论文,深度学习结合可微分光线追踪实现高保真的3D人脸重建 Title:Towards High Fidelity Monocular Face Reconstruction with Rich Reflectance using Self-supervised Learning and Ray Tracing 论文:https://arxiv.org/abs/2103.15432
  • 回复关键字:FASTLIO2
标题:港大火星实验室FAST-LIO框架大升级,实现超100Hz的LiDAR速率,支持更多LiDAR和平台! Title:FAST-LIO2: Fast Direct LiDAR-inertial Odometry 论文:https://github.com/hku-mars/FAST_LIO/blob/main/doc/Fast_LIO_2.pdf 代码:https://github.com/hku-mars/FAST_LIO
  • 回复关键字:技术总结
V技术指南--技术总结 链接:https://pan.baidu.com/s/1VnMNnZPgr_GErWqlKj_Cqg 提取码:CVCV
  • 回复关键字:DSAP
Title:IROS2021: Drone System for Autonomous Power line Inspection 标题:IROS2021,用于自主电力线路检查的无人机系统,直接挂高压电线上自动充电! 链接:https://findresearcher.sdu.dk:8443/ws/portalfiles/portal/183912681/IROS2021.pdf
  • 回复关键字:Transformer
链接:https://pan.baidu.com/s/18AQ1JQxHpmb10Cxdg17gQA 提取码:7ijg
  • 回复关键字:高速结构光
直播地址:https://app0s6nfqrg6303.h5.xiaoeknow.com/v2/course/alive/l_60fac44de4b08f7ad24018e4?app_id=app0S6nfQRG6303&alive_mode=0&pro_id=p_609281f5e4b0d4eb039721a1&type=2 直播时间:7月24日晚八点
  • 回复关键字:PCTF
Title:Pose-Controllable Talking Face Generation by Implicitly Modularized Audio-Visual Representation 标题:单张图片AI生成视频真假难辨,唇形准确同步,脑袋也能自由转动(代码开源) 论文:https://arxiv.org/abs/2104.11116 代码:https://hangz-nju-cuhk.github.io/projects/PC-AVS
  • 回复关键字:CorrNet3D
论文链接:https://arxiv.org/pdf/2012.15638.pdf 代码链接:https://github.com/ZENGYIMING-EAMON/CorrNet3D
  • 回复关键字:AS3D
Title:Active SLAM using 3D Submap Saliency for Underwater V olumetric Exploration 标题:ICRA2021,卡内基梅隆大学机器人研究所,水下环境下基于三维子图显著性的主动SLAM探测应用 论文:http://www.cs.cmu.edu/~kaess/pub/Suresh20icra.pdf
  • 回复关键字:AWUL
标题:GPS信息丢失,利用3D语义信息来精准定位自动驾驶车辆位置,精度达到10m! Title:Any Way You Look at It: Semantic Crossview Localization and Mapping With LiDAR 论文:https://ieeexplore.ieee.org/document/9361130
  • 回复关键字:CO3D
链接:https://pan.baidu.com/s/1UZhFOGhglGPWgLJ1UJXo8Q 密码:lsxx
  • 回复关键字:PTEC
标题:IROS 2021,使用事件相机对电力线路进行巡查,线路寿命增加十倍!代码开源! Title: Powerline Tracking with Event Cameras (IROS 2021) 代码:https://github.com/uzh-rpg/line_tracking_with_event_cameras
  • 回复关键字:Range Image
代码开源:https://github.com/PRBonn/range-mcl
  • 回复关键字:VIO
代码开源:https://github.com/Zumbalamambo/2D-3D-pose-tracking.
  • 回复关键字:激光
直播地址: https://app0s6nfqrg6303.h5.xiaoeknow.com/v2/course/alive/l_61414282e4b04518c6144667?app_id=app0S6nfQRG6303&alive_mode=0&pro_id=p_6118d835e4b0a27d0e3dd43c&type=2 2021年9月19日晚八点
  • 回复关键字:R3LIVE
论文地址: https://github.com/hku-mars/r3live/blob/master/papers/R3LIVE:%20A%20Robust%2C%20Real-time%2C%20RGB-colored%2C%20LiDAR-Inertial-Visual%20tightly-coupled%20stateEstimation%20and%20mapping%20package.pdf 代码:https://github.com/hku-mars/r3live
  • 回复关键字:MVSNeRF
标题:ICCV2021,MVSNeRF,基于多视图立体快速泛化的NeRF重建,效果比原始NeRF更好! Title:MVSNeRF: Fast Generalizable Radiance Field Reconstruction from Multi-View Stereo 链接:https://arxiv.org/abs/2103.15595 代码:https://github.com/apchenstu/mvsnerf
  • 回复关键字:ICRA-VDEU
Title:ICRA 2020,Vision-aided Dynamic Exploration of Unstructured Terrain 标题:ICRA 2020,MIT机器狗在杂乱地形下能跑、能跳,大幅增强机械狗的地形通过能力! 链接:https://ieeexplore.ieee.org/abstract/document/9196777
  • 回复关键字:多传感器
直播地址: https://app0s6nfqrg6303.h5.xiaoeknow.com/v2/course/alive/l_613a09d9e4b0b558b92e9b81?app_id=app0S6nfQRG6303&alive_mode=0&pro_id=&type=2 2021年9月12日晚八点
  • 回复关键字:IROS-SMED
标题:IROS2020,动态场景中的速度和内存效率高的稠密 RGB-D SLAM Title:Speed and Memory Efficient Dense RGB-D SLAM in Dynamic Scenes 链接:http://ras.papercept.net/images/temp/IROS/files/0404.pdf
  • 回复关键字:车端激光
代码:http://wiki.ros.org/velo2cam_calibration
  • 回复关键字:ICRA-SDK
标题:[ICRA2021] SDK:基于深度学习和显著性估计的点云关键点检测 Title:SKD: Keypoint Detection for Point Clouds using Saliency Estimation 链接:https://www.robots.ox.ac.uk/~mobile/drs/Papers/2021RAL_tinchev.pdf
  • 回复关键字:L3DOP
标题:ICCV2021,浙大工作!遮挡环境下基于学习的未知空间占用分布预测,机器人运动轨迹更加光滑、安全! Title:Learning-based 3D Occupancy Prediction for Autonomous Navigation in Occluded Environments 链接:https://arxiv.org/abs/2011.03981v1
  • 回复关键字:Query
标题:ICCV2021,顶尖实例分割模型:QueryInst Title:Instances as Queries 链接:https://arxiv.org/abs/2105.01928v3 代码:https://github.com/hustvl/QueryInst
  • 回复关键字:RMAN
标题:CVPR 2021,点云配准工作!循环多视图对齐网络的无监督表面点云配准! Title:Recurrent Multi-view Alignment Network for Unsupervised Surface Registration (CVPR 2021) 论文:https://arxiv.org/abs/2011.12104 代码:https://wanquanf.github.io/RMA-Net.html
  • 回复关键字:R3LIVE
标题:港大火星实验室最新工作,R3LIVE:一个健壮、实时的Lidar-IMU-视觉紧耦合顶尖SLAM框架 Title:R3LIVE: A Robust, Real-time, RGB-colored, LiDAR-Inertial-Visual tightly-coupled state Estimation and mapping package 论文:https://arxiv.org/abs/2109.07982 代码:https://github.com/hku-mars/r3live
  • 回复关键字:SimROD
Ttile:SimROD: A Simple Adaptation Method for Robust Object Detection 标题:ICCV2021,华为提出鲁棒无监督自适应目标检测方法 论文:https://arxiv.org/abs/2107.13389 代码:https://github.com/reactivetype/simrod
  • 回复关键字:GICP
标题:KITTI上准确率第二,GPU加速GICP匹配代价因子的全局一致的3D LiDAR映射 Title:Globally Consistent 3D LiDAR Mapping with GPU-accelerated GICP Matching Cost Factors 链接:https://arxiv.org/abs/2109.07073
  • 回复关键字:LineTR
Title:Line as a Visual Sentence: Context-aware Line Descriptor for Visual Localization 标题:LineTR:用于视觉定位的上下文感知线描述符 链接:https://arxiv.org/abs/2109.04753 代码:https://github.com/yosungho/LineTR
  • 回复关键字:GPA
标题:GPA:凝视增强感知系统,视觉辅助无人机安全飞行(ICRA2021) Title:GPA-Teleoperation: Gaze Enhanced Perception-aware Safe Assistive Aerial Teleoperation 链接:https://arxiv.org/abs/2109.04907
  • 回复关键字:谷歌X实习生
补充材料视频 https://share.weiyun.com/XbuFfAvcIROS 开会讲解视频 https://share.weiyun.com/R3myU8Rc 讲解视频字幕 https://share.weiyun.com/8bgNvu0U
  • 回复关键字:RVM
标题:字节跳动研究,4K图像76FPS实时抠图!代码开源! Title:Robust High-Resolution Video Matting with Temporal Guidance 论文:http://arxiv.org/abs/2108.11515 代码:https://github.com/PeterL1n/RobustVideoMatting
  • 回复关键字:国旗头像
渐变国旗图像制作工具 通道一:https://sourl.cn/8zv3Z2 通道二:https://sourl.cn/FNATTE 通道三:https://sourl.cn/LhMsd6 如果有拦截,复制上方任意地址,粘贴到手机浏览器中打开!
  • 回复关键字:CVPR-HPS
标题:[CVPR 2021] HPS:无需外部传感器,大场景、长时间的动作捕获系统 title:[CVPR 2021] HPS:3d Human Pose Estimation And Self-localization In Large Scenes From Body-mounted Sensors 论文:http://virtualhumans.mpi-inf.mpg.de/papers/guzov-mir21HPS/guzov-mir21HPS.pdf 代码:http://virtualhumans.mpi-inf.mpg.de/hps/
  • 回复关键字:DCSR
标题:[ICCV2021] 港科大:基于对准注意力模块的双摄像头超分辨率恢复 Title:[ICCV 2021 (Oral Presentation)] Dual-Camera Super-Resolution with Aligned Attention Modules 论文:https://arxiv.org/abs/2109.01349 代码:https://tengfei-wang.github.io/Dual-Camera-SR/index.html
  • 回复关键字:NALSM
Title:[SIGGRAPH 2021] Neural Animation Layering for Synthesizing Martial Arts Movements 标题:SIGGRAPH 2021,AI武术指导化!帮助动画师生成武术动作! 项目:https://github.com/sebastianstarke/AI4Animation 论文:https://github.com/sebastianstarke/AI4Animation/blob/master/Media/SIGGRAPH_2021/Paper.pdf
  • 回复关键字:LVISAM
直播地址: https://app0s6nfqrg6303.h5.xiaoeknow.com/v2/course/alive/l_616b5a51e4b05adf2075c665?app_id=app0S6nfQRG6303&alive_mode=0&pro_id=p_612106d2e4b020185a9db0fe&type=2 直播时间: 2021年10月17日晚八点。
  • 回复关键字:Choeo
标题:网易互娱:音乐驱动舞蹈动作合成(SIGGRAPH 2021) Title:SIGGRAPH 2021,ChoreoMaster : Choreography-Oriented Music-Driven Dance Synthesis 论文:https://netease-gameai.github.io/ChoreoMaster/Paper.pdf 项目:https://netease-gameai.github.io/ChoreoMaster/
  • 回复关键字:DLIO
标题:ICRA2021:基于多模态地标统一跟踪的紧耦合激光-视觉-惯性里程计 Title:Dynamic LIDAR Inertial Odometry with the Ross Robotics Extrm Robot 链接:https://arxiv.org/abs/2011.06838
  • 回复关键字:单目深度估计
单目深度估计 直播时间:2021年11月7日晚八点 直播地址:https://lgo.h5.xeknow.com/s/4sEIIM
  • 回复关键字:IGSRM
标题:[IROS2021] 卡内基梅隆大学,通过触觉感知物体重心,提升机器人抓取稳定性 Title:IROS2021-Improving Grasp Stability with Rotation Measurement from Tactile Sensing 链接:https://arxiv.org/abs/2108.00301
  • 回复关键字:YDOLO
标题:[ICCV2021] 浙大&商汤最新研究:构建时空记忆力,用于提升3D对象检测和跟踪! Title:You Don't Only Look Once: Constructing Spatial-Temporal Memory for Integrated 3D Object Detection and Tracking 代码:https://zju3dv.github.io/udolo/
  • 回复关键字:CodeMap
标题:[RA-L2021] 英国帝国理工最新工作,使用紧凑场景表示稀疏SLAM实时稠密建图! Title:CodeMapping: Real-Time Dense Mapping for Sparse SLAM using Compact Scene Representations 论文:https://arxiv.org/abs/2107.08994
  • 回复关键字:ICRA-ABTER
标题:[ICRA2021] 基于自动行为树的机械臂系统异常处理框架,针对抓取过程中的各种异常! Title:Automated Behavior Tree Error Recovery Framework for Robotic Systems 论文:https://ieeexplore.ieee.org/abstract/document/9561002
  • 回复关键字:MVTN
论文地址:https://arxiv.org/abs/2011.13244 项目地址:https://github.com/ajhamdi/MVTN
  • 回复关键字:TRIOD
标题:ICCV 2021:几行代码提升对旋转目标的检测精度,目标检测中的旋转不变性 title:Towards Rotation Invariance in Object Detection - ICCV 2021 论文:https://arxiv.org/abs/2109.13488 代码:https://github.com/akasha-imaging/ICCV2021
  • 回复关键字:CVPR2021-TCMR
Title:TCMR: Beyond Static Features for Temporally Consistent 3D Human Pose and Shape from a Video 标题:[CVPR2021] TCMR:超越静态特征的视频时间一致3D人体姿态和形状估计 论文:https://arxiv.org/abs/2011.08627 代码:https://github.com/hongsukchoi/TCMR_RELEASE
  • 回复关键字:ICCV2021-NEAT
Title:NEAT: Neural Attention Fields for End-to-End Autonomous Driving 标题:ICCV2021NEAT:端到端自动驾驶的神经注意场方法 链接:http://www.cvlibs.net/publications/Chitta2021ICCV.pdf 代码:https://github.com/autonomousvision/neat
  • 回复关键字:FVFSC
标题:[SIGGRAPH2021]快速且通用的流体-固体相互作用3D湍流仿真模拟 Title:Fast and Versatile Fluid-Solid Coupling for Turbulent Flow Simulation (SIGGRAPH Asia 2021) 论文:http://geometry.caltech.edu/pubs/LLDL21.pdf
  • 回复关键字:ADSOW
标题:空中机器人利用机载电脑感知计算躲避动态小障碍物 Title:Avoiding dynamic small obstacles with onboard sensing and computating on aerial robots 论文:https://arxiv.org/abs/2103.00406 代码:https://github.com/hku-mars/dyn_small_obs_avoidance
  • 回复关键字:FAR-Planner
标题:开源扫地机SLAM导航算法:FAR Planner全局路径规划 Title:FAR Planner: Fast, Attemptable Route Planner using Dynamic Visibility Update 代码:https://github.com/MichaelFYang/far_planner
  • 回复关键字:LEMO
标题:ICCV2021:学习 3D 场景中 4D 人体捕捉的运动先验,实现高质量三维人体动作捕捉 Title:LEMO:Learning Motion Priors for 4D Human Body Capture in 3D Scenes 链接:https://arxiv.org/pdf/2108.10399.pdf 代码:https://sanweiliti.github.io/LEMO/LEMO.html
  • 回复关键字:Refer
Title:[CVPR 2021] Refer-it-in-RGBD: A Bottom-up Approach for 3D Visual Grounding in RGBD Images 标题:[CVPR 2021] 深圳市大数据研究院:根据文本描述进行三维物体定位 链接:https://arxiv.org/abs/2103.07894v3 代码:https://unclemedm.github.io/Refer-it-in-RGBD/
  • 回复关键字:SpSeq
Title:SpSequenceNet: Semantic Segmentation Network on 4D Point Clouds 标题:[CVPR2020]SpSequenceNet:4D点云语义分割网络 链接:https://ieeexplore.ieee.org/document/9156504 代码:https://github.com/dante0shy/SpSequenceNet
  • 回复关键字:ROMO
标题:ICCV2021:ROMO:单目、单阶段、多人的3D语义分割方法 Ttile:ROMP: Monocular, One-stage, Regression of Multiple 3D People (ICCV 2021) 链接:https://arxiv.org/abs/2008.12272v4 代码:https://github.com/Arthur151/ROMP
  • 回复关键字:SFIST
标题:SIGGRAPH2021,北大&北电最新研究,模拟水面表面张力,效果自然、真实! Title:[SIGGRAPH 2021] Solid-Fluid Interaction with Surface-Tension-Dominant Contact 论文:https://lwruan.com/publication/waterstrider/
  • 回复关键字:SegICP
Code: https://github.com/jaymwong/seline
  • 回复关键字:CRNP
标题:CVPR2020:级联细化网络用于点云补全(代码开源) Title:Cascaded Refinement Network for Point Cloud Completion 链接:https://arxiv.org/abs/2004.03327v3 代码:https://github.com/xiaogangw/cascaded-point-completion
  • 回复关键字:DORVS
标题:[IROS2021] 使用镜面反射减少机器人视觉系统中的遮挡 Title:IROS2021:Diminishing Occlusions in Robot Vision Systems with Mirror Reflections 链接:https://arxiv.org/abs/2108.13599
  • 回复关键字:RoadMap
标题:华为最新工作:RoadMap:用于自动驾驶视觉定位的轻量化语义地图 Title:RoadMap: A Light-Weight Semantic Map for Visual Localization towards Autonomous Driving 链接:https://arxiv.org/abs/2106.02527v1
  • 回复关键字:DSP-SLAM
标题:代码开源,伦敦大学:DSP-SLAM:具有深度形状先验的物体语义级SLAM Title:DSP-SLAM: Object Oriented SLAM with Deep Shape Priors 链接:https://arxiv.org/abs/2108.09481 代码:https://github.com/JingwenWang95/DSP-SLAM
  • 回复关键字:DM-VIO
标题:RA-L 2022:延迟边缘化单目视觉惯性里程计,效果高于双目! Title:DM-VIO: Delayed Marginalization Visual-Inertial Odometry 链接:https://vision.in.tum.de/research/vslam/dm-vio?redirect=1 代码:https://github.com/lukasvst/dm-vio
  • 回复关键字:IROS-EKF
Title:A decentralized framework for simultaneous calibration, localization and mapping with multiple LiDAR 论文:https://arxiv.org/abs/2007.01483 硬件:https://github.com/hku-mars/lidar_car_platfrom 软件:https://github.com/hku-mars/decentralized_loam 解读:https://mp.weixin.qq.com/s/hbFCUjFR7nCKSZI2rAjw_w
  • 回复关键字:Panop
标题:ICRA2022:自适应环境的移动机器人自主导航 Title:An Efficient Locally Reactive Controller for Safe Navigation in Visual Teach and Repeat Missions 链接:https://arxiv.org/abs/2201.03938
  • 回复关键字:MLOAM
Title:Robust Odometry and Mapping for Multi-LiDAR Systems with Online Extrinsic Calibration 标题:港科大工作,在线外参标定多激光雷达的鲁棒性里程计和建图,代码开源 论文:https://arxiv.org/abs/2010.14294 代码:https://github.com/gogojjh/M-LOAM
  • 回复关键字:IROS-AIB
Title:Aerial Imagery based LIDAR Localization for Autonomous Vehicles 标题:IROS2020,基于航空影像的自动驾驶汽车激光雷达定位 链接:https://arxiv.org/abs/2003.11192
  • 回复关键字:D4LCN
原文地址:https://arxiv.org/pdf/1912.04799v1 官方代码:https://github.com/dingmyu/D4LCN
  • 回复关键字:cvpr2021
CVPR 2021 自动驾驶相关论文汇总 链接: https://pan.baidu.com/s/1Q2GqhxkkZOq2yeTreGGB2g 提取码: skgm
  • 回复关键字:CVPR21检测
CVPR2021检测类论文汇总: 链接: https://pan.baidu.com/s/1ghyDitfeGcnh1C_U2HliIA 密码: 91m8
  • 回复关键字:MonoRec
Title:[CVPR 2021] MonoRec: Semi-Supervised Dense Reconstruction in Dynamic Environments from One Camera 标题:[CVPR2021] MonoRec:无需激光雷达,单相机实现三维场景的稠密重建,去除运动伪影,泛化性能好!代码开源! 论文:https://arxiv.org/abs/2011.11814 代码:https://github.com/Brummi/MonoRec
  • 回复关键字:DLT
标题:CVPR2021,利用深度学习自动监测肿瘤病灶,准确率高达85%,效率提升14倍(数据集开源) Title:Deep Lesion Tracker: Monitoring Lesions in 4D Longitudinal Imaging Studies 链接:https://arxiv.org/abs/2012.04872 数据集:https://github.com/JimmyCai91/DLT
  • 回复关键字:三角化
直播主题:SLAM \SfM中的两视图三角化算法 直播地址:https://app0s6nfqrg6303.h5.xiaoeknow.com/v2/course/alive/l_60f0419be4b08f7ad23d3b53?app_id=app0S6nfQRG6303&alive_mode=0&pro_id=&type=2 直播时间:7月25日 周日晚上20:00
  • 回复关键字:深度估计
直播时间:7月17日晚八点 直播地址:https://app0s6nfqrg6303.h5.xiaoeknow.com/v2/course/alive/l_60eeeda8e4b08f7ad23cd33f?app_id=app0S6nfQRG6303&alive_mode=0&pro_id=p_60acfdd9e4b07e4d7fd8bcaf&type=2
  • 回复关键字:LoFTR
标题:浙大+商汤联合工作,LoFTR:利用transformers对稀疏纹理进行匹配(CVPR2021) Title:LoFTR: Detector-Free Local Feature Matching with Transformers 论文:https://arxiv.org/pdf/2104.00680.pdf 代码:https://github.com/zju3dv/LoFTR
  • 回复关键字:实时三维测量
直播时间:8月1日 上午 10:00 直播地址:https://app0s6nfqrg6303.h5.xiaoeknow.com/v2/course/alive/l_6105105ee4b054ed7c486287?app_id=app0S6nfQRG6303&alive_mode=0&pro_id=&type=2
  • 回复关键字:M3DR
标题:商汤提出手机端单目实时三维重建系统,实现逼真AR效果和交互 Title:Mobile3DRecon: Real-time Monocular 3D Reconstruction on a Mobile Phone 论文:http://www.cad.zju.edu.cn/home/gfzhang/projects/mobile-3d-recon.pdf
  • 回复关键字:ECCV-FVS
论文:ECCV2020,循环神经网络自由合成任意视角图像,代码开源!效果惊艳! Title:Free View Synthesis(ECCV2020) 链接:http://vladlen.info/papers/FVS.pdf 代码:https://github.com/intel-isl/FreeViewSynthesis
  • 回复关键字:LiDAR里程计
论文地址: https://pan.baidu.com/s/1SZgS6koRGmUsBlkpUneD8w 提取码:v2nl 代码地址:https://github.com/PRBonn/puma 视频地址:https://mp.weixin.qq.com/s/xrFAoBjC8llMcAFeD6PGMA
  • 回复关键字:EMOT多目标跟踪
Title:EagerMOT: 3D Multi-Object Tracking via Sensor Fusion 标题:ICRA2021论文,EagerMOT,通过传感器融合进行3D多目标跟踪 链接:https://arxiv.org/abs/2104.14682 代码:https://github.com/aleksandrkim61/EagerMOT
  • 回复关键字:激光雷达
直播链接:https://app0S6nfQRG6303.h5.xiaoeknow.com/v2/course/alive/l_60cb5830e4b0f120ffc83199?app_id=app0S6nfQRG6303&alive_mode=0&pro_id=&type=2 直播时间:2021年6月19日 周六 下午3-5点
  • 回复关键字:结构光
直播链接:https://app0S6nfQRG6303.h5.xiaoeknow.com/v2/course/alive/l_60d5a23ee4b057a4e7281810?app_id=app0S6nfQRG6303&alive_mode=0&pro_id=&type=2 直播时间:2021年6月26日 周六上午10点
  • 回复关键字:Transformer
链接:https://pan.baidu.com/s/18AQ1JQxHpmb10Cxdg17gQA 提取码:7ijg
  • 回复关键字:双目结构光
双目结构光,直播地址:https://app0S6nfQRG6303.h5.xiaoeknow.com/v2/course/alive/l_60aface1e4b0c7264219b522?app_id=app0S6nfQRG6303&alive_mode=0&pro_id=p_609281f5e4b0d4eb039721a1&type=2 直播时间:2021年5月30日上午10点
  • 回复关键字:PatchmatchNet
直播地址:https://app0S6nfQRG6303.h5.xiaoeknow.com/v2/course/alive/l_60a7b113e4b0c7264217e6fe?app_id=app0S6nfQRG6303&alive_mode=0&pro_id=&type=2 主题:PatchmatchNet:基于传统PatchMatch算法的高效Multi-view Stereo框架
  • 回复关键字:210625
代码:https://github.com/lijx10/DeepI2P 论文:https://arxiv.org/abs/2104.03501
  • 回复关键字:ICRA2021
链接: https://pan.baidu.com/s/1kzFsXARpoLnTnbfBmLReEQ 提取码: 68d7
  • 回复关键字:点云补全
直播地址:https://app0s6nfqrg6303.h5.xiaoeknow.com/v2/course/alive/l_60d9d38ae4b057a4e728db27?app_id=app0S6nfQRG6303&alive_mode=0&pro_id=&type=2 直播时间:7月3日 周六晚20:00
  • 回复关键字:FASTLIO
Title:FAST-LIO: A Fast, Robust LiDAR-inertial Odometry Package by Tightly-Coupled Iterated Kalman Filter 标题:香港大学工作,FAST-LIO:基于紧密耦合迭代卡尔曼滤波的快速、鲁棒的激光雷达里程计包,代码开源 链接:https://arxiv.org/pdf/2010.08196.pdf 代码:https://github.com/hku-mars/FAST_LIO
  • 回复关键字:025
课件→链接: https://pan.baidu.com/s/16mY75DfBtz0wxiDwQaG8PQ密码: xcbo 回放视频:https://www.bilibili.com/video/BV1Bf4y1b7PF/
  • 回复关键字:DELTAS
数据集和预训练模型:https://github.com/magicleap/DELTAS
  • 回复关键字:多维感知
直播链接:https://app0S6nfQRG6303.h5.xiaoeknow.com/v2/course/alive/l_60b19b02e4b07e4d7fd9efbd?app_id=app0S6nfQRG6303&alive_mode=0&pro_id=p_609d2d59e4b0fe322013088d&type=2 直播时间:2021年5月29日晚八点
  • 回复关键字:ENVGPU
GPU编程之环境搭建:Ubuntu16.04+CUDA9.0+cuDNN7.6.5+Clion 链接:https://pan.baidu.com/s/1m8BtYqluwnObfaKZRYs9-w 提取码:gn4t
  • 回复关键字:3DONE
Title: One Shot 3D Photography 标题:一张照片3D摄影 链接:https://arxiv.org/abs/2008.12298 代码:https://github.com/facebookresearch/one_shot_3d_photography 项目:https://facebookresearch.github.io/one_shot_3d_photography/ 介绍: 来自Facebook的最新研究,使用单张图像获取3D图片,这次算法真的非常接近实际应用了,拍一张照片,基于移动端优化过的神经网络,在手机端处理时间只需要几秒钟,就可以直接生成3D图片,效果惊艳(代码开源,ACM SIGGRAPH 2020)
  • 回复关键字:Geo2019
标题:使用卫星地图辅助的基于图像的地理定位算法 Title:IJCV2019: Image-Based Geo-Localization Using Satellite Imagery 链接:https://arxiv.org/abs/1903.00159v3
  • 回复关键字:RASLAM
Title:RadarSLAM: Radar based Large-Scale SLAM in ALL Weathers 标题:出自英国赫瑞-瓦特大学爱丁堡机器人技术中心,发表在最新IROS 2020上,RadarSLAM:在恶劣天气环境下基于纯雷达的SLAM解决方案 项目:http://pro.hw.ac.uk/research/radarslam/ 论文:https://arxiv.org/abs/2005.02198 数据集(恶劣天气):http://pro.hw.ac.uk/radiate/
  • 回复关键字:P-MVSNet
论文链接: https://openaccess.thecvf.com/content_ICCV_2019/papers/Luo_P-MVSNet_Learning_Patch-Wise_Matching_Confidence_Aggregation_for_Multi-View_Stereo_ICCV_2019_paper.pdf
  • 回复关键字:ASC
标题: 最新来自瑞士苏黎世联邦理工学院脚踩一对风火轮能跑能跳的轮式机器人Ascento 2,论文发表在 ICRA 2020 上! Title: LQR-Assisted Whole-Body Control of a Wheeled Bipedal Robot with Kinematic Loops 链接:https://www.ascento.ethz.ch/wp-content/uploads/2020/05/LQR_Assisted_Whole_Body_Control_for_a_Wheeled_Bipedal_Robot_with.pdf
  • 回复关键字:FDD
标题:经典3d视觉论文回顾,Facebook用『SLAM+实时光流』实现AR真实遮挡效果,发表在顶刊 ACM Transactions on Graphic 2018,代码开源 Title:Fast Depth Densification for Occlusion-aware Augmented Reality 项目:https://homes.cs.washington.edu/~holynski/publications/occlusion/index.html 论文:https://homes.cs.washington.edu/~holynski/publications/occlusion_sa2018.pdf 代码:https://github.com/facebookresearch/AR-Depth
  • 回复关键字:三维物体
论文地址: https://moscow.sci-hub.se/2490/d000c0bb4ef2d65d339a5d547b25b68e/yulanguo2014.pdf#view=FitH
  • 回复关键字:RGB检测
论文下载:https://openaccess.thecvf.com/content/ACCV2020/papers/Yan_SDCNet_Size_Divide_and_Conquer_Network_for_Salient_Object_Detection_ACCV_2020_paper.pdf 论文代码:https://github.com/Sssssbo/SDCNet (等待上传)
  • 回复关键字:Speech2Face
项目地址:https://speech2face.github.io/
  • 回复关键字:AMZ
Title:Learning-based Model Predictive Control for Autonomous Racing 标题:基于学习的自动驾驶赛车控制 论文:https://www.research-collection.ethz.ch/bitstream/handle/20.500.11850/351561/08754713.pdf?sequence=1&isAllowed=y
  • 回复关键字:地图管理
论文打包下载: https://pan.baidu.com/s/1kRX7s5GIzOjSAy1eAhi2HQ 提取码: s5ak
  • 回复关键字:ROBC
Title:RoboCut: Hot-wire Cutting with Robot-controlled Flexible Rods 标题:机器人控制的自动热线切割 链接:https://n.ethz.ch/~sduenser/content/papers/robocut/paper.pdf 来源:SIGGRAPH 2020、ACM Transactions on Graphics
  • 回复关键字:EF-Net
论文下载:https://www.sciencedirect.com/science/article/abs/pii/S0031320320305434
  • 回复关键字:Bundle Adjustment
本文所有例子代码地址: https://github.com/qxiaofan/awesome-BA-exercise
  • 回复关键字:缺陷检测
链接:https://pan.baidu.com/s/1V_aHJIgW-lbppQc0UAU_wA 提取码:y20u
  • 回复关键字:BlendedMVS
数据集和预训练模型:https://github.com/YoYo000/BlendedMVS
  • 回复关键字:3D鸟类重建
代码链接:https://github.com/marcbadger/avian-mesh 项目链接:https://marcbadger.github.io/avian-mesh 论文链接:https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123630001.pdf
  • 回复关键字:图像分割
2020年CV综述论文分方向整理-图像分割篇 链接:https://pan.baidu.com/s/1hxAu3Nlu7GAyTRIme4bQ6g 密码:y9ji
  • 回复关键字:UMM
标题:来自牛津大学,统一的多模态地标跟踪,用于紧密耦合激光雷达-视觉-惯性里程计 Title:Unified Multi-Modal Landmark Tracking for Tightly Coupled Lidar-Visual-Inertial Odometry 链接:https://arxiv.org/abs/2011.06838
  • 回复关键字:缺陷检测方法
链接:https://pan.baidu.com/s/1V_aHJIgW-lbppQc0UAU_wA 提取码:y20u
  • 回复关键字:RepVGG
代码地址: https://github.com/DingXiaoH/RepVGG
  • 回复关键字:目标检测
大盘点 | 2020年5篇目标检测算法最佳综述链接: https://pan.baidu.com/s/1IRZeDN4P8TjWu2MJIc261A 密码:3f90
  • 回复关键字:RGB图像重建
项目地址:https://smalr.is.tue.mpg.de/ 论文地址:http://files.is.tue.mpg.de/black/papers/zuffiCVPR2018.pdf
  • 回复关键字:Camera​答案
参考答案地址: https://github.com/ethan-li-coding/StereoV3DCode
  • 回复关键字:坐标系与相机参数
代码地址:StereoV3DCode:https://github.com/ethan-li-coding/StereoV3DCode
  • 回复关键字:医学影像
链接:https://pan.baidu.com/s/1xvyglXmry25XDUdarWKeDA 密码:mnbu
  • 回复关键字:VOR
标题:最新ICRA2020论文:视觉里程计回顾:什么特征应该被学习? Title: Visual Odometry Revisited: What Should Be Learnt? 论文:https://arxiv.org/pdf/1909.09803.pdf 代码:https://github.com/Huangying-Zhan/DF-VO
  • 回复关键字:U-CMR
项目链接:https://shubham-goel.github.io/ucmr/ 代码链接:https://github.com/shubham-goel/ucmr/tree/master
  • 回复关键字:CoST
Title: LION: Lidar-Inertial Observability-Aware Navigator for Vision-Denied Environments标题: LION: 用于光线恶劣环境下激光雷达+惯性SLAM框架 链接: https://arxiv.org/abs/2102.03443
  • 回复关键字:Fast-MVSNet
开源代码:https://github.com/svip-lab/FastMVSNet
  • 回复关键字:FOG
Title:Seeing Through Fog Without Seeing Fog: Deep Multimodal Sensor Fusion in Unseen Adverse Weather <img style="width:1px;height:1px;">标题:CVPR2020,在雾霾等不利天气下相机、毫米波雷达等多模态传感器深度融合 <img style="width:1px;height:1px;">代码:https://github.com/princeton-computational-imaging/SeeingThroughFog
  • 回复关键字:Python视觉实战项目
链接:https://pan.baidu.com/s/1NhoKLMzjpsE-SJNct1RxfQ 提取码:noob
  • 回复关键字:Deep Learning
链接:https://pan.baidu.com/s/16yDuVfhppKreQ78e6qqD5A 提取码:vsdm
  • 回复关键字:ODMD
Title: Depth from Camera Motion and Object Detection 标题:从相机运动和目标检测中获取深度(CVPR2021) 论文:https://arxiv.org/abs/2103.01468 代码:https://github.com/griffbr/ODMD
  • 回复关键字:视频特效
视频特效 (蚂蚁呀嘿)链接: https://pan.baidu.com/s/1ozsSFsLTgv9DPVhoCHGZjA 密码: o36k
  • 回复关键字:PMS
Windows版本:https://github.com/ethan-li-coding/PatchMatchStereo Linux版本:https://github.com/Haonan-DONG/PatchMatchStereo_linux
  • 回复关键字:20210302
论文:https://vincentqin.gitee.io/paper-collection/20210302/2006.12567v2.pdf 项目:https://github.com/changhao-chen/deep-learning-localization-mapping
  • 回复关键字:MRF
标题:CVPR2021,基于内关联性的单目实时全身动作捕获,速率达到25帧/秒 Title:CVPR2021, Monocular Real-time Full Body Capture with Inter-part Correlations 论文:https://arxiv.org/abs/2012.06087 简介:来自清华大学工作!本文提出了一种用于实时全身捕获的方法,该方法可以从单色图像估计身体和手的形状和运动以及动态3D面部模型,使用了一种新的神经网络架构,该架构以高计算效率利用了身体与手部之间的相关性,在多个数据集上达到了State-of-art!
  • 回复关键字:TRA
Title: Track to Detect and Segment: An Online Multi-Object Tracker 标题:跟踪目标检测和分割:在线多对象跟踪器(CVPR2021,代码开源) 项目:https://jialianwu.com/projects/TraDeS.html
  • 回复关键字:CVPR2021 Oral
CVPR2021 Oral论文合集 链接:https://pan.baidu.com/s/1Cm_SUTjyfFcTZDDHIbG4eg 提取码:miu6
  • 回复关键字:CenterPoint
代码:https://github.com/tianweiy/CenterPoint 论文:https://arxiv.org/pdf/2006.11275.pdf
  • 回复关键字:RPR
Title: Robust Place Recognition using an Imaging Lidar 标题:最新ICRA2021论文,使用成像激光雷达的稳健位置识别 论文、代码:https://github.com/TixiaoShan/imaging_lidar_place_recognition
  • 回复关键字:CDFI
Title:CDFI: Compression-Driven Network Design for Frame Interpolation 标题:用于帧插值的压缩驱动网络设计 代码:https://github.com/tding1/CDFI
  • 回复关键字:CenterNet2
代码:https://github.com/xingyizhou/CenterNet2 论文:https://arxiv.org/abs/2103.07461
  • 回复关键字:DEFMO
Title:DeFMO: Deblurring and Shape Recovery of Fast Moving Objects 标题:最新CVPR2021论文,DeFMO网络:快速运动物体的去模糊和形状恢复,代码开源 论文:https://arxiv.org/abs/2012.00595 代码:https://github.com/rozumden/DeFMO
  • 回复关键字:LIT
标题:LiTAMIN2:基于激光雷达的超轻型SLAM,处理速度最快达到1000HZ(ICRA2021) Title:LiTAMIN2: Ultra Light LiDAR-based SLAM using Geometric Approximation applied with KL-Divergence 链接:https://arxiv.org/abs/2103.00784
  • 回复关键字:鲁棒点云
论文代码:https://github.com/fukexue/RGM.
  • 回复关键字:monorec
[CVPR 2021]: 动态场景单目相机半监督密集重构论文 链接: http://xxx.itp.ac.cn/pdf/2011.11814.pdf
  • 回复关键字:PISE
项目主页:http://cic.tju.edu.cn/faculty/likun/projects/PISE 代码:https://github.com/Zhangjinso/PISE
  • 回复关键字:RRDL
Title:Online LiDAR-SLAM for Legged Robots with Robust Registration and Deep-Learned Loop Closure 标题:ICRA2020论文,基于CNN和点云分割的LiDAR-SLAM点云回环检测方法,用于四足ANYmal机器人导航 链接:https://ori.ox.ac.uk/media/5601/2020icra_ramezani.pdf
  • 回复关键字:EGO
Title: EGO-Swarm: A Fully Autonomous and Decentralized Quadrotor Swarm System in Cluttered Environments 标题: 最新ICRA2021论文,EGO-Swarm:在杂乱环境中的全自动分散式四旋翼群系统 论文: https://arxiv.org/abs/2011.04183v1 代码: https://github.com/ZJU-FAST-Lab/ego-planner-swarm
  • 回复关键字:0322
文章链接:http://arxiv.org/pdf/2103.04273 代码链接:https://github.com/ChenyangLEI/flash-reflection-removal
  • 回复关键字:SpinNet
论文代码:https://github.com/QingyongHu/SpinNet
  • 回复关键字:OOD
Title:Online Object Searching by a Humanoid Robot in an Unknown Environment 标题:最新ICRA2021论文,人形机器人在未知环境下自主寻找物体,代码开源! 链接:https://ieeexplore.ieee.org/document/9361266 代码:https://github.com/TsuruMasato/OnlineObjectDetector
  • 回复关键字:FLOAM
标题:经典SLAM框架LOAM升级版!Fast LOAM,速度提升3倍! 链接:https://github.com/wh200720041/floam
  • 回复关键字:qd3dt
itle:Monocular Quasi-Dense 3D Object Tracking 标题:单目准稠密3D目标追踪,代码开源! 链接:https://arxiv.org/abs/2103.07351 代码:https://github.com/SysCV/qd-3dt
  • 回复关键字:R2LIVE
Title:R2LIVE: A Robust, Real-time, LiDAR-Inertial-Visual tightly-coupled state Estimator and mapping 标题:R2LIVE:稳健,实时,LiDar、惯性、相机的多传感器紧密耦合SLAM框架(代码开源) 链接:https://arxiv.org/abs/2102.12400 代码:https://github.com/hku-mars/r2live
  • 回复关键字:PVST
Title:PVStereo: Pyramid Voting Module for End-to-End Self-Supervised Stereo Matching 标题:最新ICRA2021论文,PVStereo: 基于金字塔投票模块的端到端的自监督立体匹配 论文:https://arxiv.org/abs/2103.07094
  • 回复关键字:相机定位
论文地址:https://arxiv.org/pdf/2103.16792.pdf Code开源:https://github.com/Tangshitao/DenseScene-Matching
  • 回复关键字:NeX
Title:NeX: Real-time View Synthesis with Neural Basis Expansion 标题:CVPR2021论文,NEX:基于神经基扩展的实时视图合成,速度快1000倍,代码开源! 链接:https://arxiv.org/pdf/2103.05606.pdf 代码:https://github.com/nex-mpi/nex-code/
  • 回复关键字:生成3D场景
链接:https://pan.baidu.com/s/1J63PSdeARGgIktFtKSdgew 密码:4dq9
  • 回复关键字:FFB
Title:FFB6D: A Full Flow Bidirectional Fusion Network for 6D Pose Estimation 标题:最新CVPR2021论文,FFB6D:用于6D位姿估计的全流双向融合网络 论文:https://arxiv.org/abs/2103.02242 代码:https://github.com/ethnhe/FFB6D
  • 回复关键字:rosbag
代码仓库:https://github.com/AbangLZU/multi_lidar_calibration rosbag 下载链接:https://pan.baidu.com/s/1Ny3Ji9aF1ZTjgklhxPvDbg 提取码: pemc
  • 回复关键字:GNMS
标题:GrooMeD-NMS:用于单目3D目标检测的分组微分NMS,代码开源! Title:GrooMeD-NMS: Grouped Mathematically Differentiable NMS for Monocular 3D Object Detection 链接:https://arxiv.org/pdf/2103.17202.pdf 代码:https://github.com/abhi1kumar/groomed_nms/
  • 回复关键字:GBTE
Title:Graph-Based Topological Exploration Planning in Large-Scale 3D Environments (ICRA 2021) 标题:无人机探洞!大规模三维环境中无人机基于图的拓扑探索规划(ICRA2021) 链接:https://arxiv.org/abs/2103.16829
  • 回复关键字:MVSNet网络
开源代码(tensorflow):https://github.com/YoYo000/MVSNet 开源代码(PyTorch):https://github.com/XYZ-qiyh/MVSNet_pytorch
  • 回复关键字:SSL
Title:SSL_SLAM: Lightweight 3-D Localization and Mapping for Solid-State LiDAR 标题:SSL_SLAM: 使用固态雷达的轻量级3D定位和映射(代码开源) 链接:https://arxiv.org/abs/2102.03800 代码:https://github.com/wh200720041/ssl_slam
  • 回复关键字:LVI-SAM
代码开源:http://github.com/TixiaoShan/LVI-SAM
  • 回复关键字:PHOSA
项目网址:https://jasonyzhang.com/phosa/ 论文网址:https://arxiv.org/pdf/2007.15649.pdf
  • 回复关键字:OMNI
Title:OmniPhotos: Casual 360° VR Photography 标题:OmniPhotos:360°VR全景照片生成,目前最快,代码开源 论文:https://richardt.name/publications/omniphotos/OmniPhotos-BertelEtAl-SIGAsia2020-paper.pdf 代码:https://github.com/cr333/OmniPhotos 项目:https://richardt.name/publications/omniphotos/
  • 回复关键字:PSN
Title:PlaneSegNet: Fast and Robust Plane Estimation Using a Single-stage Instance Segmentation CNN 标题:PlaneSegNet:使用单阶段实例分割CNN的快速鲁棒平面估计(CVPR2021) 链接:https://arxiv.org/abs/2103.15428
  • 回复关键字:PAConv
论文链接:https://arxiv.org/abs/2103.14635 代码链接:https://github.com/CVMI-Lab/PAConv
  • 回复关键字:colmap
colmap论文地址: https://openaccess.thecvf.com/content_cvpr_2016/html/Schonberger_Structure-From-Motion_Revisited_CVPR_2016_paper.html
  • 回复关键字:ECML
Title:Extrinsic Calibration of Multiple LiDARs of Small FoV in Targetless Environments 标题:自动驾驶中多个小视场固态激光雷达不依赖参照物的外参标定(代码即将开源) 论文:https://ieeexplore.IEEE.org/document/9361153 代码:https://github.com/hku-mars/multi_lidar_calib 解读:https://zhuanlan.zhihu.com/p/353243916
  • 回复关键字:PREDATOR
论文代码:https://github.com/ShengyuH/OverlapPredator
  • 回复关键字:ICRA SLAM
链接:https://pan.baidu.com/s/1JUEGAl94qnRQLWKsTrSjNA 提取码:4xa8
  • 回复关键字:DLAQ
Title:Dynamic Landing of an Autonomous Quadrotor on a Moving Platform in Turbulent Wind Conditions 标题:ICRA2020,多旋翼无人机移动平台快速自动降落算法,航母可用!出自麻省理工航天控制实验室 论文:http://www.lewissoft.com/pdf/ICRA2020/1956.pdf
  • 回复关键字:kimera
Title:Kimera: an Open-Source Library for Real-Time Metric-Semantic Localization and Mapping 标题:MIT工作:Kimera,新的实时度量语义的开源SLAM系统(ICRA2020) 论文:https://arxiv.org/abs/1910.02490 代码:https://github.com/MIT-SPARK/Kimera
  • 回复关键字:CenterPoint
直播地址:https://app0S6nfQRG6303.h5.xiaoeknow.com/v2/course/alive/l_6096a173e4b09890f0ea05fe?app_id=app0S6nfQRG6303&alive_mode=0&pro_id=&type=2 CenterPoint:三维点云目标检测算法梳理及最新进展(CVPR2021)
  • 回复关键字:FDVD
Title:FAST-Dynamic-Vision: Detection and Tracking Dynamic Objects with Event and Depth Sensing 标题:有了这个算法,你的无人机再也不怕撞鸟了,IROS2021最新论文,代码开源! 链接:https://arxiv.org/abs/2103.05903v2 代码:https://github.com/ZJU-FAST-Lab/FAST-Dynamic-Vision
  • 回复关键字:pixelNeRF
项目地址:https://alexyu.net/pixelnerf/ 论文地址:https://arxiv.org/pdf/2012.02190.pdf
  • 回复关键字:ERLE
标题:ICRA2021,基于 LiDAR 的高效、鲁棒深度学习端到端的自动驾驶导航控制,大幅减少接管次数(代码开源) Title:Efficient and Robust LiDAR-Based End-to-End Navigation 论文:https://arxiv.org/abs/2105.09932 代码:https://le2ed.mit.edu/ 地址:https://www.youtube.com/watch?v=5ZyhHxjgUGo
  • 回复关键字:OMNI
Title:OmniDet: Surround View Cameras based Multi-task Visual Perception Network for Autonomous Driving 标题:ICRA2021,OmniDet:用于自动驾驶的基于环视摄像头的多任务视觉感知网络 链接:https://arxiv.org/abs/2102.07448 代码:https://github.com/valeoai/WoodScape 解读:https://mp.weixin.qq.com/s/5mqOk7gAmOiIGB7qCo3rJQ
  • 回复关键字:Intensity-SLAM
论文地址:https://ieeexplore.ieee.org/document/9354903 代码地址:https://github.com/wh200720041/intensity_slam
  • 回复关键字:F4D
标题:Function4D: 从稀疏的消费级RGB-D相机中精确重建人体动作(CVPR2021) Title:Function4D: Real-time Human Volumetric Capture from Very Sparse Consumer RGBD Sensors 链接:http://www.liuyebin.com/Function4D/Function4D.html
  • 回复关键字:单目自监督
代码地址:https://github.com/nvlabs/umr 论文地址:https://arxiv.org/pdf/2003.06473.pdf
  • 回复关键字:NFRA
标题:CVPR2021,NormalFusion:实时高分辨率 RGB-D表面重建 Title:NormalFusion: Real-Time Acquisition of Surface Normals for High-Resolution RGB-D Scanning 链接:http://www.minhkim.org/cvpr2021p3/cvpr2021-normalfusion-main-paper.pdf
  • 回复关键字:3DVideo
标题:CVPR2021,不需要训练数据,CNN就可以稳定视频?不可思议! Title:3D Video Stabilization with Depth Estimation by CNN-based Optimization 链接:https://drive.google.com/file/d/1vTalKtMz2VEowUg0Cb7nW3pzQhUWDCLA/view 代码:https://yaochih.github.io/deep3d-stabilizer.io/
  • 回复关键字:VSNET
Title:VS-Net: Voting with Segmentation for Visual Localization (CVPR 2021) 标题:CVPR2021论文,VS-Net:基于语义投票的视觉定位,代码即将开源! 链接:https://arxiv.org/abs/2105.10886
  • 回复关键字:AVPS
Title:AVP-SLAM: Semantic Visual Mapping and Localization for Autonomous Vehicles in the Parking Lot 标题:AVP-SLAM:停车场自动泊车SLAM框架,采用停车线等作为定位标志,IROS2020 论文:https://arxiv.org/abs/2007.01813
  • 回复关键字:LiTAMIN2
视频链接:https://www.bilibili.com/video/BV1yp4y1h792/ 论文链接: https://arxiv.org/pdf/2103.00784v1.pdf
  • 回复关键字:PESC
Title:Pixel-level Extrinsic Self Calibration of High Resolution LiDAR and Camera in Targetless Environment 标题:代码开源,不使用外部标定物的高分辨率LiDAR和相机的像素级外部自标定 链接:https://arxiv.org/abs/2103.01627 代码:https://github.com/hku-mars/livox_camera_calib
  • 回复关键字:AOGTS
标题:强化学习来学习超跑自动驾驶,跑得比人还快!老司机了,人类又在一个领域输给了AI,ICRA2021 Title:Autonomous Overtaking in Gran Turismo Sport Using Curriculum Reinforcement Learning 论文:很抱歉,由于链接检查原因,无法附送,请按标题搜索论文地址
  • 回复关键字:DRO
论文链接:https://arxiv.org/abs/2103.13201 代码链接:https://github.com/aliyun/dro-sfm
  • 回复关键字:逆相机法
链接:https://pan.baidu.com/s/1SaYMPHj3Mh_A9mQIGx-R8g 提取码:803z
  • 回复关键字:二维图像GAN
原文链接:https://openreview.net/pdf?id=FGqiDsBUKL0 代码地址:https://github.com/XingangPan/GAN2Shape
  • 回复关键字:RSSS
标题:CVPR2021,基于多视点鱼眼图像的实时球面扫描立体成像 Title:Real-Time Sphere Sweeping Stereo from Multiview Fisheye Images(CVPR2021) 链接:http://www.minhkim.org/cvpr2021p1/cvpr21-sphere-sweeping.pdf
  • 回复关键字:导航系统
论文地址: https://pan.baidu.com/s/1YutBKtSNDNe8FN8BX8Untw 提取码:wc0k 视频地址:https://mp.weixin.qq.com/s/9dPmQ6pd8-vRg8NFv_nNPA
  • 回复关键字:kangle
DIY一条自己的机械狗,硬件图纸开源 链接:https://pan.baidu.com/s/1fRzGUFes1-Db6G0NlLIkMg 提取码:4kac
  • 回复关键字:6D姿态
论文及代码地址: https://geometry.stanford.edu/projectsCS_CVPR2019/
  • 回复关键字:CenterFusion
论文链接: https://arxiv.org/pdf/2011.04841.pdf 源码链接: https://github.com/mrnabati/CenterFusion
  • 回复关键字:半直接单目SLAM
【代码地址】https://github.com/sunghoon031/LCSD_SLAM
  • 回复关键字:ACCV2020
链接: https://pan.baidu.com/s/14fYkk-Pm9YRJMQF4Uw8-Og 提取码: mxqv
  • 回复关键字:SR2020
标题:基于强化学习,发表在Science子刊,自主学会爬山的四足机器人!算法开源! 论文:Learning Quadrupedal Locomotion over Challenging Terrain 地址:https://arxiv.org/abs/2010.11251v1 项目:https://leggedrobotics.github.io/rl-blindloco/ 代码:https://github.com/leggedrobotics/learning_quadrupedal_locomotion_over_challenging_terrain_supplementary
  • 回复关键字:CasMVSNet
论文代码:https://github.com/alibaba/cascade-stereo
  • 回复关键字:视觉和惯性传感器
论文链接:https://pan.baidu.com/s/1AsYnlQXAFGextg9f78C75g 提取码:f8yk 相关视频和代码:https://Smilels.github.io/multimodal-translation-teleop
  • 回复关键字:FASTER
IEEE IROS 2019,最新无人机飞控算法,在未知环境下的无人机快速安全航迹规划(代码开源) 论文:FASTER: Fast and Safe Trajectory Planner for Flights in Unknown Environments 链接:https://arxiv.org/abs/1903.03558v3 代码:https://github.com/mit-acl/faster 模拟:https://github.com/jtorde
  • 回复关键字:Monodepth2
论文链接:https://sci-hub.se/10.1109/iccv.2019.00393
  • 回复关键字:PyTorch教程
链接:https://pan.baidu.com/s/1aEukpjCFe6XdE15QBKWezQ 提取码:re61
  • 回复关键字:SIL
Title:Optimization Based Motion Planning for Multi-Limbed Vertical Climbing Robots 标题:基于优化运动规划器的六足垂直攀爬机器人(荣获IROS 2019 安全救援领域最佳论文) 链接:https://arxiv.org/abs/1909.06339?context=cs.RO 简介:这个机器人能爬墙,在IROS 2019上荣获“安全,保卫和救援机器人最佳论文奖”!SILVIA(具有智能铰接功能的六足机器人)是一种六足机器人,可以在基于优化的运动计划器的帮助下在非平坦的墙壁间攀爬,向上攀爬并避开障碍物!
  • 回复关键字:MADER
Title:MADER: Trajectory Planner in Multi-Agent and Dynamic Environments 标题:出自麻省理工航天控制实验室,最新无人机飞控算法,动态环境下快速避障(2020, 代码即将开源) 论文:https://arxiv.org/abs/2010.11061 代码:https://github.com/mit-acl/mader
  • 回复关键字:NSVF
Title:Neural Sparse Voxel Fields 标题:来自Facebook的最新研究,发表在NeuraIPS 2020,基于神经网络自由视点场景快速、高质量渲染,效果大幅领先现有算法(代码开源)! 链接:https://arxiv.org/abs/2007.11571 代码:https://github.com/facebookresearch/NSVF
  • 回复关键字:三维卷积
论文:https://arxiv.org/pdf/2004.02803.pdf 代码:https://github.com/XinyiYing/D3Dnet
  • 回复关键字:DOF
Title:Generating Grasp Poses for a High-DOF Gripper Using Neural Networks 标题:基于神经网络的高自由度机械臂抓取 链接:https://arxiv.org/abs/1903.00425v4
  • 回复关键字:改进的点对特征
论文下载地址: https://dacemirror.sci-hub.se/proceedingsarticle/6c04e927e78ba7ce954a4b4a490dc1ac/vidal2018.pdf#view=FitH
  • 回复关键字:Matlab
Linux下:Matlab R2016a 链接: https://pan.baidu.com/s/12aR5cC2ysg07WakwxNBt0A 提取码: fpuh Windows: Matlab2018a 链接: https://pan.baidu.com/s/1lPN5_3UZvBEG2iJX0ApxTw 提取码: it9t
  • 回复关键字:Anaconda
Anaconda 百度云下载链接:https://pan.baidu.com/s/17hV6y5DygUNV7dIb3xt67A 密码:fd0j
  • 回复关键字:解AX=XB
手眼标定核心方程:AX=XB解法参考论文: 链接: https://pan.baidu.com/s/1N41lR-Ja4_YiBfQJZi-0Uw 提取码: nrmt
  • 回复关键字:PCL1.8.1+VS2015
1、PCL-1.8.1-All-In-One; 2、对应pdb 3、源码 链接:https://pan.baidu.com/s/1XW_C4cN5EdhQPUcYQnNu7A 提取码:kgsm
  • 回复关键字:PCL-1.8.1源码
PCL-1.8.1源码下载: 链接: https://pan.baidu.com/s/1tLKZ-i03XwxVbdsZ-LLGWQ 提取码: srfx 链接若失效,请加小助理微信,备注:获取PCL-1.8.1源码资料。
  • 回复关键字:直通滤波
直通滤波相关代码与效果图: 链接: https://pan.baidu.com/s/1xhKKNdfziXh61l9NB3i2OQ 提取码: bhmx
  • 回复关键字:体素法滤波
体素法滤波代码+效果图: 链接: https://pan.baidu.com/s/1Hi8dGVwrv965pMqd_KZoog 提取码: cytr
  • 回复关键字:统计滤波
统计滤波相关代码与效果图: 链接: https://pan.baidu.com/s/1cQA0jCA6P8uSpFZKDkQqng 提取码: fvp3
  • 回复关键字:投影滤波
链接:https://pan.baidu.com/s/1MpVOFszzOLzvSsk7gGXMnA 提取码:8v72
  • 回复关键字:Linux
Linux资料: 链接:https://pan.baidu.com/s/1NnO7LBpmm1oszZYBQcQs7g 提取码:ta0r
  • 回复关键字:立体视觉综述
StereoVIsion原文: 链接:https://pan.baidu.com/s/1BEj1T35xCVPhlYm6zgH4Xw 提取码:1s81
  • 回复关键字:OpenCL
OpenCL学习资料: 链接:https://pan.baidu.com/s/1xs6iA4tFw0SCi5ynQ468kA 提取码:zwoc
  • 回复关键字:Git
链接:https://pan.baidu.com/s/1AG8Uf8GQg2BU_3no1Hb4NQ 提取码:i7mi
  • 回复关键字:Shell
链接:https://pan.baidu.com/s/17KcJASA_AxRSfvg6_qYleg 提取码:ll9h
  • 回复关键字:数学基础
数学基础: 链接:https://pan.baidu.com/s/1JMgvGU179ueoJ6DLSOaesw 提取码:14d3
  • 回复关键字:数据结构与算法
数据结构与算法: 链接:https://pan.baidu.com/s/1gJZNjxoIzvfARrg5bIXd6A 提取码:zxsm
  • 回复关键字:C++
链接:https://pan.baidu.com/s/14jUIMKMWckqaD6w2_26nGw 提取码:m5vv
  • 回复关键字:Python
Python资料: 链接:https://pan.baidu.com/s/1iAqSLeIfA2HsUuPjej6B6A 提取码:vf9a
  • 回复关键字:C语言
C语言: 链接:https://pan.baidu.com/s/1GwY9xAjHzrFTK2Y9-kWupw 提取码:i3t3
  • 回复关键字:ROS
ROS资料: 链接:https://pan.baidu.com/s/1-79odnoLvRODG5rSu-09kA 提取码:b6mn
  • 回复关键字:深度学习
链接: https://pan.baidu.com/s/12fljXusaWhh8E6aqU0MgQQ 提取码: if2w
  • 回复关键字:计算机视觉
计算机视觉: 链接:https://pan.baidu.com/s/17_xzUY2hgKb0jpQxNG3zoQ 提取码:la3w
  • 回复关键字:立体视觉
链接:https://pan.baidu.com/s/1zGi_NIg9lZyIWgY1TJvj1w 提取码:56bg
  • 回复关键字:事件相机作者
作者微信:larry_dongy
  • 回复关键字:DSO
DSO学习资料: 链接: https://pan.baidu.com/s/1L8v0ZODbvRNaZNNLwD8Tsw 提取码: 86vz
  • 回复关键字:DirectBA

链接:https://pan.baidu.com/s/1B78E7WTwv_4ZLBVAKcSoBQ 提取码:bu9a

  • 回复关键字:CVPR2020_Event_Camera
链接: https://pan.baidu.com/s/1Q5D6yaF7PZk2_JnpndpoNg 提取码: 9umj
  • 回复关键字:LRNNet
论文下载:https://arxiv.org/pdf/2006.02706.pdf
  • 回复关键字:位姿识别论文
基于三维向量对的乱序堆叠物体的位姿识别 ,参考论文: 链接: https://pan.baidu.com/s/1O3Hba4cqhEO7NYgY1lkkqg 提取码: pxrw
  • 回复关键字:3D视觉优质源码
源码分享:https://t.zsxq.com/qN3FyZb 格雷码:https://t.zsxq.com/zf6iyvV 相机标定:https://t.zsxq.com/yzfmeIU 缺陷检测::https://t.zsxq.com/YvjuF6M 深度估计与深度补全:https://t.zsxq.com/EiQvRNb SLAM:https://t.zsxq.com/aYR3JMv 点云:https://t.zsxq.com/U7qrJYz 立体匹配:https://t.zsxq.com/3v3vFAY 注:该模块需付费,星球成员可免费学习,由星主与合伙人以及各嘉宾持续完善中。
  • 回复关键字:基于结构光的立体视觉
链接:https://pan.baidu.com/s/1KLMXPQEpELToxoZGKwUkGQ 提取码:pm2r
  • 回复关键字:3D视觉github资源汇总
CV相关的会议与期刊:https://github.com/qxiaofan/awesome_3d_conference_journals
CV公司:https://github.com/qxiaofan/awesome_3d_vision_company_summary
深度学习:https://github.com/Tom-Hardy-3D-Vision-Workshop/awesome-deep-learning
三维重建:https://github.com/qxiaofan/awesome_3d_restruction SLAM:https://github.com/qxiaofan/awesome_slam_papers
6D姿态估计:https://github.com/Tom-Hardy-3D-Vision-Workshop/awesome-6D-pose-estimation
点云处理:https://github.com/qxiaofan/awesome_PointCloud_process
深度估计:https://github.com/Tom-Hardy-3D-Vision-Workshop/awesome-depth-estimation
自动驾驶:https://github.com/Tom-Hardy-3D-Vision-Workshop/awesome-Autopilot-algorithm
3D检测:https://github.com/Tom-Hardy-3D-Vision-Workshop/awesome-3D-object-detection
缺陷检测:https://github.com/Tom-Hardy-3D-Vision-Workshop/awesome-defect-detection
3D视觉精品文章汇总:https://github.com/qxiaofan/awesome-3D-Vision-Papers
  • 回复关键字:CS231n
CS231n: 链接:https://pan.baidu.com/s/1ahVVqG3z1ar6epXbmd6MHQ
  • 回复关键字:ECCV2020
ECCV2020所有论文打包下载: 链接: https://pan.baidu.com/s/1LATSUmTQXRGquIcRb0M64g 提取码: g3a2
  • 回复关键字:机器人装配姿态估计
机器人装配姿态估计:链接:https://pan.baidu.com/s/1sI8iyQaDNAEmsYdefN4Pag 提取码:cbtw
  • 回复关键字:PPF物体检测
PPF物体检测:链接:https://pan.baidu.com/s/1_PenNtf5dKbkPUaYjW92SA 提取码:yjhh
  • 回复关键字:CenterPoint
链接:https://pan.baidu.com/s/1QN1KF88-dhVjFcBSE1tAsw 提取码:qxd2
  • 回复关键字:ICP
链接:https://pan.baidu.com/s/1G4ct7Z2_Sgw_fUztj7q5mg 提取码:1w0d
  • 回复关键字:silhonet
链接:https://pan.baidu.com/s/1Z8R25Uul2PUKuR6BYTFI6Q 提取码:xg84
  • 回复关键字:V-LOAM
链接:https://pan.baidu.com/s/1ELiXkCloG5IAM3wglMUkOA 提取码:f0ud
  • 回复关键字:视频防抖
视频防抖 代码:https://github.com/spmallick/learnopencv/tree/master/VideoStabilization 素材:链接: https://pan.baidu.com/s/1gQLnqOZeNevzOpLifEpuMw 提取码: jbay
  • 回复关键字:6D对象姿势估计
论文地址:https://arxiv.org/abs/2001.01869

回复关键字:Late-Fusion

论文下载地址:https://arxiv.org/pdf/2009.00784.pdf
  • 回复关键字:点云可视化
链接:https://pan.baidu.com/s/17VApJ59CmXgPi410nfWrVg 提取码:kzil
  • 回复关键字:红外
《红外目标检测结果收集》链接:https://pan.baidu.com/s/1aiIjl-aLJ9iHnBrnkDbnPA 提取码:pdcv
  • 回复关键字:6D姿态估计
论文地址:http://arxiv.org/abs/1911.08324v2
  • 回复关键字:3D目标检测
原文链接: https://arxiv.org/pdf/2007.12392.pdf
  • 回复关键字:Supervoxels
链接:https://pan.baidu.com/s/1du6AQ0jsxsUwzDn-pW0-WA 提取码:0jqp
  • 回复关键字:点云配准
论文下载地址: 链接:https://pan.baidu.com/s/1yx2wmqV3GaN4-uTOurwn9Q 提取码:xt77
  • 回复关键字:计算机视觉算法与应用
链接:https://pan.baidu.com/s/1OUU356iUziyakRSLCsAoXQ 提取码:c4ca
  • 回复关键字:双目事件相机
论下载链接:https://pan.baidu.com/s/1xoSlfeyaejBCTXyluPw-wQ 提取码:sceb
  • 回复关键字:LOAM相关论文
下载链接: https://pan.baidu.com/s/1AWWUHGn48v2Dde1YizIvlg 提取码: rmcw
  • 回复关键字:自动标注
论文下载:https://pan.baidu.com/s/1zFFgQcT6fC6-ZspBLB7irw 提取码:9dbm
  • 回复关键字:SOLO
论文地址:v1: https://arxiv.org/abs/1912.04488 ,v2: https://arxiv.org/abs/2003.10152。
  • 回复关键字:深度学习地图定位
论文地址:https://arxiv.org/pdf/2006.12567.pdf
  • 回复关键字:SuperGlue
项目:https://psarlin.com/superglue/ 论文:https://arxiv.org/abs/1911.11763 代码:https://github.com/magicleap/SuperGluePretrainedNetwork 解读:https://vincentqin.gitee.io/posts/superglue/#comments
  • 回复关键字:R-MVSNet
开源数据集下载链接: https://pan.baidu.com/s/1Wb9E6BWCJu4wZfwxm_t4TQ#list/path=%2F 提取码:s2v2
参考论文所在地址: 链接:https://pan.baidu.com/s/1uNc57QiG4wq3kJyA8mz72Q 提取码:v3a9
  • 回复关键字:工业相机
链接:https://pan.baidu.com/s/1Viq36d3HAnVASyb8LPQxDg 提取码:53sa
  • 回复关键字:三维目标识别
论文下载链接:https://pan.baidu.com/s/1RZyyZGPGX7ZwD4A82i9FAw 提取码:9fb8
  • 回复关键字:路径规划
论文:Motion Primitives-based Agile Exploration Path Planning for Aerial Robotics 代码:https://github.com/unr-arl/mbplanner_ros
  • 回复关键字:CenterNet
论文下载:https://arxiv.org/abs/1904.07850 代码下载:https://github.com/Qjizhi/TensorRT-CenterNet-3D
  • 回复关键字:PIFuHD
欢迎大家关注我们的视频号,找到对应的视频,下方评论区给出了 Paper: https://arxiv.org/pdf/2004.00452.pdf Code: https://github.com/facebookresearch/pifuhd
  • 回复关键字:Robot Learning
参考文献三篇文章下载链接: https://pan.baidu.com/s/1OZ4Z2Wtv2UeQJrB5P1R7zA 提取码:44ly
  • 回复关键字:Atlas
project: http://zak.murez.com/atlas/ paper: https://arxiv.org/abs/2003.10432 code: https://github.com/magicleap/Atlas
  • 回复关键字:opencv
opencv书籍链接: https://pan.baidu.com/s/1pUtzAxZo841RZqb2z6IcHQ 提取码:vf3q
各版本 opencv链接: https://pan.baidu.com/s/1D6NHlVsUjWCJ35I2Y9KVVA 提取码:52CV
  • 回复关键字:colmap
colmap论文地址: https://openaccess.thecvf.com/content_cvpr_2016/html/Schonberger_Structure-From-Motion_Revisited_CVPR_2016_paper.html
  • 回复关键字:人脸识别技术
人脸识别技术:链接:https://pan.baidu.com/s/18Z2eIRBevqF9k5_o7TTdNg 提取码:w5lz
  • 回复关键字:图像配准算法
论文下载地址:https://dacemirror.sci-hub.se/journal-article/22b507c65c1a2b127afa06ed312b828b/guo2014.pdf#view=Fit 或者百度网盘下载:链接:https://pan.baidu.com/s/1YF5wIJ6KpjXJDoROv6lcFg 提取码:ob0a
  • 回复关键字:HP-2019
链接:https://arxiv.org/abs/1901.03798 代码:https://github.com/chanyn/3Dpose_ssl.git
  • 回复关键字:MoreFusion
论文:<MoreFusion: Multi-object Reasoning for 6D Pose Estimation from Volumetric Fusion> 链接:https://arxiv.org/abs/2004.04336 代码:暂未开源 推荐:https://github.com/j96w/DenseFusion(其它工作)
  • 回复关键字:PackNet
代码:https://github.com/tri-ml/packnet-sfm
  • 回复关键字:ECCV-PIV
链接:https://pan.baidu.com/s/1o1gTq1u67rG3UT9UDfIycA 提取码:2e73
  • 回复关键字:RPM-NET
Paper: https://arxiv.org/abs/2003.13479 Code: https://github.com/yewzijian/RPMNet
  • 回复关键字:CVPR2020-CRN
论文:https://arxiv.org/abs/2004.03327 代码:https://github.com/xiaogangw/cascaded-point-completion
  • 回复关键字:IROS2019-SLAM
链接:https://pan.baidu.com/s/1D48p_XEhbYb6emBIVmjX8w 提取码:2v62
  • 回复关键字:ECCV2020-FEG
Title:Flow-edge Guided Video Completion(ECCV2020) 标题:最新ECCV2020论文:基于光流的视频修复算法,效果大幅超越之前算法!(代码开源) 论文:https://arxiv.org/abs/2009.01835 代码:https://github.com/vt-vl-lab/FGVC
  • 回复关键字:loam_livox
代码下载地址: https://github.com/hku-mars/loam_livox
  • 回复关键字:vid2vid
标题:Video-to-Video Synthesis 论文:https://arxiv.org/pdf/1808.06601.pdf 代码:https://github.com/NVIDIA/vid2vid
  • 回复关键字:BANet
论文下载:https://arxiv.org/pdf/2004.14582 论文代码:https://github.com/zzhanghub/bianet (等待上传)
  • 回复关键字:IROS2019-Re
论文:IROS'19-ReFusion: 3D Reconstruction in Dynamic Environments for RGB-D Cameras Exploiting Residuals 链接:https://arxiv.org/abs/1905.02082v3 代码:https://github.com/PRBonn/refusion
  • 回复关键字:NRGB-D, MEGA
NRGB-D: 代码:https://github.com/NVlabs/neuralrgbd 论文:https://arxiv.org/pdf/1901.02571.pdf
NRGB-D: 代码:https://github.com/NVlabs/neuralrgbd 论文:https://arxiv.org/pdf/1901.02571.pdf
  • 回复关键字:计算机视觉应用
参考文献下载地址: 链接:https://pan.baidu.com/s/1l1E5GwXNscbCoVD0mWcoKA 提取码:yghz
  • 回复关键字:3DP
标题:3D Photography using Context-aware Layered Depth Inpainting 链接:https://arxiv.org/abs/2004.04727v3 代码:https://github.com/vt-vl-lab/3d-photo-inpainting 工程:https://shihmengli.github.io/3D-Photo-Inpainting/ 解读:https://zhuanlan.zhihu.com/p/133289298
  • 回复关键字:CVDE
标题:Consistent Video Depth Estimation 论文:https://arxiv.org/abs/2004.15021v2 代码:https://github.com/facebookresearch/consistent_depth 解读:https://www.jianshu.com/p/c9e1e8a81c4f
  • 回复关键字: 移动最小二乘法
链接:https://pan.baidu.com/s/1DFdxdRlGqXLfdHjbVt2jsg 提取码:bp32
  • 回复关键字:9076
第三版164页手推笔记:链接: https://pan.baidu.com/s/1-G-dq6FAA7zyLbKW0ZWmbw 提取码: 8jkj
  • 回复关键字:李中伟
链接:https://pan.baidu.com/s/1bqUuyT0McrzoZtiil5SUBQ 提取码:lzw3
回复关键字:SYOLO
链接: https://pan.baidu.com/s/19TTwrvNEHBpkU8c9m99y7w 提取码: bcms
  • 回复关键字:全景视频超分
论文链接: https://arxiv.org/pdf/2008.10320.pdf](https://arxiv.org/pdf/2008.10320.pdf 参考代码:https://github.com/lovepiano/SMFN_For_360VSR](https://github.com/lovepiano/SMFN_For_360VSR
  • 回复关键字:nanodet
nanodet 代码下载链接:链接: https://pan.baidu.com/s/1QJ8SttTxGy4IVc0hzTymgQ 提取码: mget
  • 回复关键字:PVA-MVSNet
论文链接:https://arxiv.org/pdf/1912.03001.pdf
  • 回复关键字:ICCV2019
论文地址:https://node1.chrischoy.org/data/publications/fcgf/fcgf.pdf代码地址:https://github.com/chrischoy/FCGF
  • 回复关键字:车道线检测
代码:https://github.com/cfzd/Ultra-Fast-Lane-Detection
  • 回复关键字:mvsnet
mvsnet测试数据集下载
链接: https://pan.baidu.com/s/1wqXXpNmibTO5vH7WXZqpNQ 提取码: 5hw4
  • 回复关键字:SLAM学习资料
链接:https://pan.baidu.com/s/1HkwKJFmakvdjGIKdO51NCQ
提取码:n4an
  • 回复关键字:oldphoto
http://xxx.itp.ac.cn/pdf/2004.09484.pdf
  • 回复关键字: 语音导航
论文:https://arxiv.org/pdf/2004.02857.pdf 项目: https://jacobkrantz.github.io/vlnce/?fbclid=IwAR2VO1jwjaq4Uydz2O25ZaLXVFjoD46QirYnW1zNeNAJyNkleA0KS_PDBrE 代码: https://github.com/jacobkrantz/VLN-CE
  • 回复关键字:影响因子
链接:https://pan.baidu.com/s/1qW_vkzCXo6Gt-4QFiwPvzg 提取码:rnr7
  • 回复关键字:CVPR2020
链接:https://pan.baidu.com/s/1s7cDUDGKVlOlmHQjk9pIzg 提取码:gwvg
  • 回复关键字:机器学习
链接:https://pan.baidu.com/s/1lGINq-ynw9mzZW4TUZGfJA
提取码:an48
  • 回复关键字:123
Pytorch资料:链接: https://pan.baidu.com/s/1xQSv9312ETJODxdIZX69mw 提取码: 8ka3

About Joyk


Aggregate valuable and interesting links.
Joyk means Joy of geeK