论文笔记——Semantic Scene Completion from a Signal Depth Image
2018-01-09 22:50
519 查看
没有太多时间好好翻译了,大概把要点梳理了一下啦~
semantic labels for a scene from a signal-view depth map observation
以前的类似工作呢一直是将1、2任务分开完成的,但作者设计了一个end-to-end 3D Conv net:SSCNet用于完善语义场景,发现效果更好。
input:a single depth image
output:occupancy and semantic labels
Joint model
*Solution:*A dilation-based 3D context module用于扩大感受野,能够有效捕捉来自三维体积数据的信息,其中信号是稀疏且缺乏高频细节
RGB-D datasets only provide annotationos on visible surfaces,how do we obtain training data with complete volumetric annatations at scene level?
*Solution:*SUNCG datasts(we can compute 3D secen volumes with dense object labels through voxelization)
(RGB-D仅仅提供了浅层表面的注释)
提出了一个新的数据集SUNCG,该数据集是dense注释的合成三维场景数据集。
相关的算法都是建立在相机模型的基础之上,分单目(即单视角)和双目(即多视角,multi-view),而本论文是single-view。
Task:
produce a complete 3D voxel representation of volumetric occupancysemantic labels for a scene from a signal-view depth map observation
以前的类似工作呢一直是将1、2任务分开完成的,但作者设计了一个end-to-end 3D Conv net:SSCNet用于完善语义场景,发现效果更好。
input:a single depth image
output:occupancy and semantic labels
如何解决Semantic Scene Completion的思路:
A dilation-based 3D context moduleJoint model
Problems and Solution:
How do we effective capture contextual information from 3D volumetric data,where the signal is sparse and lacks high frequency detail?*Solution:*A dilation-based 3D context module用于扩大感受野,能够有效捕捉来自三维体积数据的信息,其中信号是稀疏且缺乏高频细节
RGB-D datasets only provide annotationos on visible surfaces,how do we obtain training data with complete volumetric annatations at scene level?
*Solution:*SUNCG datasts(we can compute 3D secen volumes with dense object labels through voxelization)
(RGB-D仅仅提供了浅层表面的注释)
Attribute:
提出SSCNet,该网络的输入是单幅的深度图像,输出包括了体素和语义标签。提出了一个新的数据集SUNCG,该数据集是dense注释的合成三维场景数据集。
相关的算法都是建立在相机模型的基础之上,分单目(即单视角)和双目(即多视角,multi-view),而本论文是single-view。
相关文章推荐
- Semantic Scene Completion from a Single Depth Image
- 论文笔记-深度估计(1)Depth Map Prediction from a Single Image using a Multi-Scale Deep Network
- 深度学习论文笔记--Depth Map Prediction from a Single Image using a Multi-Scale Deep Network
- [深度学习论文笔记][Depth Estimation] Depth Map Prediction from a Single Image using a Multi-Scale Deep Netw
- 【论文笔记】Depth Map Prediction from a Single Image using a Multi-Scale Deep Network
- Attribute2Image --- Conditional Image Generation from Visual Attributes 论文笔记
- 论文笔记:Show and Tell Lessons learned from the 2015 MSCOCO Image Captioning Challenge
- 论文笔记:Concept Mask: Large-Scale Segmentation from Semantic Concepts
- 论文笔记:Image Captioning with Semantic Attention
- 【论文笔记】Semantic Parsing on Freebase from Question-Answer Pairs
- Attention to Scale: Scale-Aware Semantic Image Segmentation论文阅读笔记
- 【论文笔记】Deep Visual-Semantic Alignments for Generating Image Description
- 【论文笔记】Show and Tell: Lesson learned from the 2015 MSCOCO Image Captioning Challenge
- [深度学习论文笔记][Image to Sentence Generation] Deep Visual-Semantic Alignments for Generating Image Descri
- [论文笔记]SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation
- 深度学习论文(八)---DeepLabV1-SEMANTIC IMAGE SEGMENTATION WITH DEEP CONVOLUTIONAL NETS AND FULLY CONNECTED C
- ICCV2017论文“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks” 阅读笔记
- 全卷积(FCN)论文阅读笔记:Fully Convolutional Networks for Semantic Segmentation
- 【论文笔记】 R2CNN: Rotational Region CNN for Orientation Robust Scene Text Detection
- StackGAN: Text to Photo-realistic Image Synthesis with Stacked Generative Adversarial Networks 论文笔记