Image retrieval using BIM and features from pretrained VGG network for indoor localization

Ha, Inhae; Kim, Hongjo; Park, Somin; Kim, Hyoungkwan

doi:10.1016/j.buildenv.2018.05.026

Cited by 103 publications

(50 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, even those broken images were able to be matched with 3d-images. This is because the feature maps of the 4th pooling layer of VGG extract structural characteristics rather than detailed characteristics [8].…”

Section: Discussionmentioning

confidence: 99%

“…Studies have been suggested to perform image retrieval using CNNs [5,6,7,8]. This paper used the method suggested by [8] for cross-domain image retrieval, in which the feature maps produced by the VGG network were vectorized and the vectorized feature maps were compared using the cosine distance. Feature maps are the output from a layer of a CNN and the input to the next layer of the CNN.…”

Section: Similarity Comparisonmentioning

confidence: 99%

“…Feature maps are the output from a layer of a CNN and the input to the next layer of the CNN. Following the suggestions of [8], this study, for the comparison purpose, used the 4 th pooling layer of VGG16 and VGG19 [2].…”

Section: Similarity Comparisonmentioning

confidence: 99%

See 2 more Smart Citations

Deep Learning-based Vehicle Image Matching for Flooding Damage Estimation

Park¹,

Baek²,

Sohn³

et al. 2019

Proceedings of the Creative Construction Conference 2019

Self Cite

View full text Add to dashboard Cite

Images representing flooding damages can provide valuable information, such as the damage location and severity. Automated and quantifiable analyses of those images allow asset managers to accurately understand the vulnerability of the infrastructure. To this end, this paper proposes a methodology to match a vehicle in a flooding image to a 3D vehicle image. The proposed method is a part of a framework for flooding depth estimation. As the initial step of the framework, the proposed method uses Mask R-CNN and VGG network to extract the vehicle object and its features, respectively. The features of the vehicle images are compared with those of 3D vehicle image, to find a good match. A total of 87 vehicle objects were used to validate the proposed method, and promising levels of matching accuracy were obtained. Once the framework is completed, the proposed method is expected to automatically analyze flooding images for its damage assessment.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Similarity Comparisonmentioning

confidence: 99%

See 1 more Smart Citation

Deep Learning-based Vehicle Image Matching for Flooding Damage Estimation

Park¹,

Baek²,

Sohn³

et al. 2019

Proceedings of the Creative Construction Conference 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…The use of a virtual environment in achieving a collaborative design process was supported. Case study research in [4] entailed that the collaborative design process can be conducted successfully using a shared workspace, where participants work together using one medium. Bosch-Sijtema et al [13] revealed findings that strengthen the need for a shared workspace.…”

Section: Second Map: Case-study-based Researchmentioning

confidence: 99%

“…Previous studies have developed the e-negotiation model for collaboration practices in the design process [3,4], but research on negotiation in BIM is rare. The developments of e-negotiation in virtual collaboration include automated negotiation, a virtual design studio [3], knowledge management for decision support, and a shared workspace [4]. E-negotiation is also being used and developed by AEC consulting firms in Indonesia.…”

Section: Introductionmentioning

confidence: 99%

BIM and E-Negotiation Practices in AEC Consulting Businesses

2019

View full text Add to dashboard Cite

Architectural, engineering, and construction (AEC) consulting businesses are developing into a specialized industry, and collaborative decision making is essential to obtain the finest design. Everything is now virtual. Building information modelling (BIM) has the most potential to support current practices. BIM is promising for effective and efficient design processes. Negotiation and decision-making processes appear to be activities that are the most difficult for BIM to facilitate. Both activities are hard to complete in person and virtually. Every participant has their own preferences, intention, and pay-off optimum, and conflicts are difficult to avoid. Communication and e-negotiation are main issues in BIM practices. This paper proposes BIM and e-negotiation practices in AEC consulting businesses, with the main intention to reveal critical success factors that enhance the utilization of BIM in supporting communication and e-negotiation. A survey through observation and questionnaire distribution was used to collect the data. Descriptive analysis through a mean and standard deviation scatter plot was used to analyse the data. Two hundred and two respondents consisting of 91 design managers (leaders) and 111 designers/consultants (non-leaders) were involved in the research. Based on the analysis, it was found that there are different perceptions between those two groups. The design managers stated that they were the most vital factor in supporting the e-negotiation in BIM, whereas the consultants (designers) stated that job description was the main essential factor.

show abstract

Sport training action correction by using convolutional neural network

Shi

Jiang

2020

Internet Technology Letters

View full text Add to dashboard Cite

This paper proposes a framework for correcting the actions during sport training. We estimate the positions of bone joint points and the postures by using a convolutional neural network. The positions of bone joint points and the postures are the premise to correct the actions during sport training. In convolutional neural network, it first predicts the positions of bone joint points via confidential map of different parts; then utilizes affinity domain, which is a 2‐dimensional vector set, to predict the directions and positions of four limbs; lastly repeats the above steps and analyzes the bone joint points via greedy analytical reasoning algorithm to construct the human skeleton. The personal action is represented as the constructed skeleton which is used to correct the incorrect or non‐standard actions during sports. The experimental results evaluate the effective of the proposed sport training action correction method.

show abstract

Image retrieval using BIM and features from pretrained VGG network for indoor localization

Cited by 103 publications

References 18 publications

Deep Learning-based Vehicle Image Matching for Flooding Damage Estimation

Deep Learning-based Vehicle Image Matching for Flooding Damage Estimation

BIM and E-Negotiation Practices in AEC Consulting Businesses

Sport training action correction by using convolutional neural network

Contact Info

Product

Resources

About