SegNet

新颖实用的深度全卷积神经网络结构——SegNet

全卷积神经网络结构VGG16网络
  57

Contributor

contributed at 2020-03-01

Authorship

Homepage:  
View
Is authorship not correct? Feed back

Classification(s)

Method-focused categoriesData-perspectiveIntelligent computation analysis

Model Description

English {{currentDetailLanguage}} English

Quote from: https://arxiv.org/abs/1511.00561

Present a novel and practical deep fully convolutional neural network architecture for semantic pixel-wise segmentation termed SegNet. This core trainable segmentation engine consists of an encoder network, a corresponding decoder network followed by a pixel-wise classification layer. The architecture of the encoder network is topologically identical to the 13 convolutional layers in the VGG16 network. The role of the decoder network is to map the low resolution encoder feature maps to full input resolution feature maps for pixel-wise classification. The novelty of SegNet lies is in the manner in which the decoder upsamples its lower resolution input feature map(s). Specifically, the decoder uses pooling indices computed in the max-pooling step of the corresponding encoder to perform non-linear upsampling. This eliminates the need for learning to upsample. The upsampled maps are sparse and are then convolved with trainable filters to produce dense feature maps. Compare our proposed architecture with the widely adopted FCN and also with the well known DeepLab-LargeFOV, DeconvNet architectures. This comparison reveals the memory versus accuracy trade-off involved in achieving good segmentation performance.

SegNet was primarily motivated by scene understanding applications. Hence, it is designed to be efficient both in terms of memory and computational time during inference. It is also significantly smaller in the number of trainable parameters than other competing architectures. Also performed a controlled benchmark of SegNet and other architectures on both road scenes and SUN RGB-D indoor scene segmentation tasks. Show that SegNet provides good performance with competitive inference time and more efficient inference memory-wise as compared to other architectures.

Model Metadata

Name {{metadata.overview.name}}
Version {{metadata.overview.version}}
Model Type {{metadata.overview.modelType}}
Model Domain
{{domain}}
Sacle {{metadata.overview.scale}}

There is no overview about this model. You can click to add overview.

Purpose {{metadata.design.purpose}}
Principles
{{principle}}
Incorporated Models
{{incorporatedModel}}
Model part of larger framework: {{metadata.design.framework}}
Incorporated Models
{{process}}

There is no design info about this model. You can click to add overview.

Information {{metadata.usage.information}}
Initialization {{metadata.usage.initialization}}
Hardware Requirements {{metadata.usage.hardware}}
Software Requirements {{metadata.usage.software}}
Inputs
{{input}}
Outputs
{{output}}

There is no usage info about this model. You can click to add overview.

How to Cite

Vijay Badrinarayanan (2020). SegNet, Model Item, OpenGMS, https://geomodeling.njnu.edu.cn/modelItem/0168bf76-747d-4ab4-a7c0-3078be10fb6a
Copy

History

Last modifier : 
Xu Kai
Last modify time : 
2020-12-18
Modify times : 
View History

QR Code

Contributor(s)

Initial contribute: 2020-03-01

Authorship

Homepage:  
View
Is authorship not correct? Feedback

History

Last modifier : 
Xu Kai
Last modify time : 
2020-12-18
Modify times : 
View History

QR Code

×

{{curRelation.overview}}
{{curRelation.author.join('; ')}}
{{curRelation.journal}}









You can link related {{typeName}} from repository to this model item, or you can create a new {{typeName.toLowerCase()}}.

Related Items
Related Items

You can link resource from repository to this model item, or you can create a new {{typeName.toLowerCase()}}.

Drop the file here, orclick to upload.
Select From My Space
+ add

These authorship information will be submitted to the contributor to review.

Cancel Submit
Model Classifications
Cancel Submit
Localizations + Add
{{ item.label }} {{ item.value }}
Model Name :
Cancel Submit
Name:
Version:
Model Type:
Model Domain:
Scale:
Purpose:
Principles:
Incorporated models:

Model part of

larger framework

Process:
Information:
Initialization:
Hardware Requirements:
Software Requirements:
Inputs:
Outputs:
Cancel Submit
Title Author Date Journal Volume(Issue) Pages Links Doi Operation
Cancel Submit
Add Cancel

{{articleUploading.title}}

Authors:  {{articleUploading.authors[0]}}, {{articleUploading.authors[1]}}, {{articleUploading.authors[2]}}, et al.

Journal:   {{articleUploading.journal}}

Date:   {{articleUploading.date}}

Page range:   {{articleUploading.pageRange}}

Link:   {{articleUploading.link}}

DOI:   {{articleUploading.doi}}

Yes, this is it Cancel

The article {{articleUploading.title}} has been uploaded yet.

OK
Cancel Confirm