My personal paper/code/dataset archives about computer vision
| # | Paper | Published in | Summary | Misc |
|---|---|---|---|---|
| 1 | Learning Priors for Semantic 3D Reconstruction | ECCV 2018 | 也适用于室内;暂时未开源 | |
| 2 | RayNet: Learning Volumetric 3D Reconstruction with Ray Potentials | CVPR 2018 | 基于multi-view的考虑，在learning-based重建的基础上加入了几何约束 | 非语义重建；已开源 |
| 3 | Robust Dense Mapping for Large-Scale Dynamic Environments | ICRA 2018 | SLAM + semantic segmentation + scene flow | 双目;已开源 |
| 4 | Semantic Multi-view Stereo: Jointly Estimating Objects and Voxels | CVPR 2017 | 引入shape prior解决遮挡，进行probabilistic重建 | |
| 5 | Semantically Informed Multiview Surface Refinement | ICCV 2017 | 3D网格几何表面和语义标签的联合优化 | 已开源 |
| 6 | Semantic 3D reconstruction with continuous regularization and ray potentials using a visibility consistency constraint | CVPR 2016 | ||
| 7 | Discrete Optimization of Ray Potentials for Semantic 3D Reconstruction | CVPR 2015 | 直接优化三维模型的投影和图像的误差 | |
| 8 | Joint 3D Scene Reconstruction and Class Segmentation | CVPR 2013 | 提出了一个分割和重建联合优化解决的框架 | |
| 9 | An overview of recent progress in volumetric semantic 3D reconstruction | ICPR 2016 | 语义三维重建的综述 | |
| 10 | Dense Semantic 3D Reconstruction | TPAMI 2017 | 同8 | |
| 11 | Large-Scale Outdoor 3D Reconstruction on a Mobile Device | CVIU 2017 | 移动设备（Google Project Tango)上的大规模重建 | 基本real-time；没有结合语义 |
| 12 | Incremental Dense Semantic Stereo Fusion for Large-Scale Semantic Scene Reconstruction | ICRA 2015 | 第一个实时的大范围室外场景稠密语义重建 | 总结并比较了了大规模语义重建的相关论文 |
| 13 | Large-Scale Semantic 3D Reconstruction: An Adaptive Multi-Resolution Model | CVPR 2016 | 提出了一种自适应的multi-resolution的框架 | 离线处理 |
| 14 | Joint Semantic Segmentation and 3D Reconstruction from Monocular Video | ECCV 2014 | 构建体素块的离散图重建街景 | 单目，不需要估计深度 |
| 15 | The Semantic Paintbrush: Interactive 3D Mapping and Recognition in Large Outdoor Spaces | CHI 2015 | 眼镜 + 双目RGB-红外 + 手持激光笔重建室外场景 | 实时 |
| 16 | Incremental Dense Multi-modal 3D Scene Reconstruction | IROS 2015 | stereo + lidar | 实时 |
| 17 | Urban 3D Semantic Modelling Using Stereo Vision | ICRA 2013 | ||
| 18 | Mesh Based Semantic Modelling for Indoor and Outdoor Scenes | CVPR 2013 | 提出了三维物体（网格模型）标签的生成方法 | |
| 19 | Efficient 3-d scene analysis from streaming data | ICRA 2013 | 提出了一种场景表达方式保证速度和精度 | 实时 |
| 20 | Multi-Label Semantic 3D Reconstruction using Voxel Blocks | 3DV 2016 | 利用voxel blocks解决了多标签内存占用过多的问题 | |
| 21 | Semantic 3D Reconstruction with Finite Element Bases | arXiv 2017 | 其实没看懂。。。 | |
| 22 | Joint Optimization for Object Class Segmentation and Dense Stereo Reconstruction | IJCV 2012 | 双目重建+分割联合优化框架 | |
| 23 | Extracting 3D Scene-Consistent Object Proposals and Depth from Stereo Images | ECCV 2012 | 引入了物体和场景的约束 | |
| 24 | Semantic Structure From Motion with Points, Regions, and Objects | CVPR 2012 | SfM中引入语义信息 | |
| 25 | MVSNet: Depth Inference for Unstructured Multi-view Stereo | ECCV 2018 | 从多视角图像估计深度 | 已开源 |
| 26 | 3D modeling on the go: Interactive 3d reconstruction of large-scale scenes on mobile devices | 3DV 2015 |
可以关注一下几个组和个人的主页
- ETH Computer Vision and Geometry group
- Max Planck Research Group for Autonomous Vision
- Autonomous Vision Group(和上面的是同一个组)
- Christian Häne
- Nikolay Savinov
- Patrick Pérez
| # | Paper | Published in | Summary | Misc |
|---|---|---|---|---|
| 1 | 3D Semantic Parsing of Large-Scale Indoor Spaces | CVPR 2016 | 大规模室内场景3D语义解析 | 提供了数据集 |
| 2 | Dense 3D Semantic Mapping of Indoor Scenes from RGB-D Images | ICRA 2014 | 基于贝叶斯更新和三维CRF的2D到3D标签迁移 | 实时 |
| 3 | A fully end-to-end deep learning approach for real-time simultaneous 3D reconstruction and material recognition | ICRA 2017 | 一个端到端的三维重建+材质识别系统 | 基本实时 |
| 4 | Database-Assisted Object Retrieval for Real-Time 3D Reconstruction | Computer Graphics Forum 2015 | ||
| 5 | ScanComplete: Large-Scale Scene Completion and Semantic Segmentation for 3D Scans | CVPR 2018 | 输入不完整场景预测完整模型并进行语义分割 | 已开源 |
| 6 | SemanticFusion: Dense 3D semantic mapping with convolutional neural networks | ICRA 2017 | SLAM提供匹配点，CNN进行语义预测并融合 |
重点关注
| # | Paper | Published in | Summary | Misc |
|---|---|---|---|---|
| 1 | Dense Object Reconstruction with Semantic Priors | CVPR 2013 | 通过目标检测和形状先验引入语义信息，克服了传统MVS的缺点 | 有多视角图片训练出物体的语义先验 |
| 2 | Segment Based 3D Object Shape Priors | CVPR 2015 | 提出了新的shape prior formulation，将物体分成多个凸块,重建问题可以看作容积式的多标签分割 | |
| 3 | Class Specific 3D Object Shape Priors Using Surface Normal | CVPR 2014 | 利用物体表面的法向信息作为shape prior | |
| 4 | Learning a Multi-view Stereo Machine | NIPS 2017 | 一个端到端的网络，输入多视角图片，输出Voxel Occupancy Grid或者深度图 | 已开源 |
| 5 | Semantic Object Reconstruction via Casual Handheld Scanning | SIGGRAPH ASIA 2018 | 利用语义标签提升重建质量，并提出了一种主动式的自学习框架 |
| # | Paper | Published in | Summary | Misc |
|---|---|---|---|---|
| 1 | Multi-view 3D Entangled Forest For Semantic Segmentation and Mapping | ICRA 2018 | 在语义建图中引入3DEF分类器，提出了新的多视角融合方法改善了分割效果 | |
| 2 | CNN-SLAM: Real-time dense monocular SLAM with learned depth prediction | CVPR 2017 | CNN预测单目深度并进行语义分割 | |
| 3 | REMODE: Probabilistic, Monocular Dense Reconstruction in Real Time | ICRA 2014 | 单目图片流中估计深度概率图 | |
| 4 | Quadtree-accelerated Real-time Monocular Dense Mapping | IROS 2018 | 单目实时估计深度+稠密建图 | 已开源 |
| 5 | Multi-level mapping: Real-time dense monocular SLAM | ICRA 2016 | 四叉树增强局部纹理 | |
| 6 | REMODE: Probabilistic, Monocular Dense Reconstruction in Real Time | ICRA 2014 | 已开源 |
重点关注
| # | Paper | Published in | Summary | Misc |
|---|---|---|---|---|
| 1 | Building Rome in a Day | ICCV 2009 | large scale | |
| 2 | Real-time 3D Reconstruction at Scale using Voxel Hashing | TOG 2013 | 利用voxel hashing节省GPU资源 | RGB-D;已开源 |
| 3 | Scalable Real-time Volumetric Surface Reconstruction | TOG 2013 | octree | |
| 4 | A Volumetric Method for Building Complex Models from Range Images | SIGGRAPH 1996 | 提出了volumetric式重建 |
| # | Paper | Published in | Summary | Misc |
|---|---|---|---|---|
| 1 | Direction Matters: Depth Estimation with a Surface Normal Classifier | CVPR 2015 | 在depth estimation中引入了表面法向量分类 | |
| 2 | Efficient Large-Scale Stereo Matching | ACCV 2010 | 寻找可以可靠匹配的特征点并以此为支撑做三角变换,对视差进行插值计算 | 效果一般，勉强实时；已开源（CPU） |
| 3 | Massively Parallel Multiview Stereopsis by Surface Normal Diffusion | ICCV 2015 | 已开源 |
| # | Paper | Published in | Summary | Misc |
|---|---|---|---|---|
| 1 | Visual Odometry and Mapping for Autonomous Flight Using an RGB-D Camera | ISRR 2011 |
| # | Paper | Published in | Summary | Misc |
|---|---|---|---|---|
| 1 | On Visual Similarity Based 3D Model Retrieval | 2003 | 比较silhouette的相似度 | |
| 2 | DeepCCFV: Camera Constraint-Free Multi-View Convolutional Neural Network for 3D Object Retrieval | AAAI 2019 | ||
| 3 | Multi-view Convolutional Neural Networks for 3D Shape Recognition | ICCV 2015 | ||
| 4 | Camera Constraint-Free View-Based 3-D Object Retrieval | TIP 2012 | ||
| 5 | A new 3D model retrieval approach based on the elevation descriptor |
| # | Dataset | Misc |
|---|---|---|
| 1 | 3D Warehouse | 有一些建筑模型,格式为skp，同时关注下SketchUp |
| 2 | ShapeNet | 有一些建筑（8000+）和vegetation的模型，格式为skp（数据集来源于1。。。） |
| 3 | GrabCAD | 建筑、制造、机械 |
| 4 |
| Dataset | Hyperlink | Paper | Comment |
|---|---|---|---|
| NYU Depth Dataset V2 | https://cs.nyu.edu/~silberman/datasets/nyu_depth_v2.html | RGB + D + class labels | |
| Semantic3D | http://www.semantic3d.net/ | Semantic3D: A new Large-scale Point Cloud Classification Benchmark | 大规模点云分类的benchmark |
| Cityscapes Dataset | https://www.cityscapes-dataset.com/ | 城市街道的语义理解 |