论文收集方法
检索了所有带depth的标题的论文,争取没有遗漏的筛选深度估计相关论文
下面进行分类
单纯有监督深度估计
- uncertainty aware cnns for depth completion: uncertainty from beginning to end
有监督深度补全,给出了不确定性度量 - generating and exploiting probabilistic monocular depth estimation
通用的有监督深度估计 - structure-guided ranking loss for single Image depth prediction
有监督深度估计,提出了更好的loss - online depth learning against forgetting in monocular videos
实时深度学习搞定深度估计
多任务有监督深度估计
- normal assisted stereo depth estimation
有监督学习,联合估计法线和深度提升深度估计精度 - predicting sharp and accurate occlusion boundaries in monocular depth estimation using displacement fields
有监督单目深度估计,同时预测位移域和深度图,使得边缘更好看 - SDC-depth,semantic divide-and-conquer network for monocular depth estimation
深度语义联合有监督估计 - D3VO deep depth pose and deep uncertainty for monocular visual odometry
联合深度估计,位姿估计和不确定性估计的网络 - towards better generalization joint depth pose learning without posenet
深度位姿联合估计 - the edge of depth explicit constraints between segmentation and depth
深度语义联合估计
自监督深度估计
- Bi3D Stereo Depth estimation via binary classifications
提出了Bi3D网络,通过多个二元分类器,实现双目输入下的视差估计,达到实时,从粗到精任意精度的分类效果。属于利用双目的自监督深度估计方法 - 3d packing for self supervised monocular depth estimation
从视频中的自监督深度恢复方案PackNet,预测深度,预测位姿,然后联合监督。另外提出了一个室外RGBD数据集。 - self supervised monocular trained depth estimation using self attention and discrete disparity volume
自监督深度估计,利用自注意力机制,离散分类网络视差图估计估计深度,并且能够产生不确定性 - on the uncertainty of self-supervised monocular depth estimation
自监督深度估计,同时估计不确定度 - novel view synthesis of dynamic scenes with globally coherent depths from a monocular camera
时序RGB序列的深度估计
比较特殊的深度估计
- focus on defocus: bridging the synthetic to real domain gap for depth estimation
有监督深度估计,在虚拟数据集上训练,并利用额外的虚焦监督项来做真实的深度估计 - BiFuse monocular 360 depth estimation via Bi-projection fusion
对全景视频做深度估计 - geometric structure based and Regularized depth estimation from 360° indoor imagery
全景图像下的深度估计,利用网络预测几何结构,然后利用几何结构进行深度估计,另外提出了一个全景图像的仿真数据集 - domain decluttering : simplifying images to mitigate synthetic real domain shift and improve depth estimation
域迁移:将仿真数据集的深度估计结果迁移到真实深度估计场景 - accurate estimation of body height from a single depth image via a four-stage developing network
基于深度图的精确人体高度测量 - depth sensing beyond lidar range
超长距离的深度估计 - 3D photography using context aware layered depth inpainting
使用RGBD图像预测被遮挡区域的深度和纹理 - self supervised human depth estimation from monocular videos
自监督人体深度估计 - joint graph based depth refinement and normal estimation
深度调精 - channel attention based iterative residual learning for depth map super-resolution
DSR深度图超分辨率恢复 - from depth what can you see? depth completion via auxiliary image reconstruction
稀疏深度稠密化,没有RGB输入,这里利用稀疏深度猜一个灰度图出来,然后做稀疏深度稠密化 - RoutedFusion:Learning real-time depth map fusion
利用深度学习做多个深度图的融合 - cost volume pyramid based depth inference for multi view stereo
基于RGB的重建系统
3D目标检测
- learning depth guided convolutions for monocular 3D object detection
提出了D4LCN网络,用于解决基于RGB图像的3D目标检测,传统方法是先估计深度,然后再利用深度图做基于点云的目标检测。本文直接基于RGB和估计的深度图队RGB做3D目标检测。 - handvoxnet: deep voxel based network for 3D hand shape and pose estimation from a single depth map
使用3D卷积对仅有深度图输入的手势进行姿态识别 - IDA-3D instance depth aware 3D object detection from stereo vision for autonomous driving
利用双目进行3D目标检测 - A2dele: adaptive and attentive depth distiller for efficient RGBD salient Object Detection
利用知识蒸馏,进行基于RGBD的目标检测 - 3DV:3D dynamic voxel for action Recognition in depth video
利用纯深度视频流进行动作识别
其他
- rethinking depthwise separable convolutions: how intra kernel correlations lead to improved mobileNets
CNN架构改进,提出了一个改进型的逐深度的分离卷积方案,在mobilenet上提升效果很好 - sideways depth-parallel training of video models
提出了一个针对视频的深度学习网络的并行训练方法 - deep spatial gradient and temporal depth learning for face anti-spoofing
人脸识别反欺诈
更多推荐
2020CVPR深度估计
发布评论