Skip to main content

Shallow2Deep: Indoor scene modeling by single image understanding.

Nie, Y., Guo, S., Chang, J., Han, X., Huang, J., Hu, S.M. and Zhang, J. J., 2020. Shallow2Deep: Indoor scene modeling by single image understanding. Pattern Recognition, 103 (July), 107271.

Full text available as:

2002.09790.pdf - Accepted Version
Available under License Creative Commons Attribution Non-commercial.


DOI: 10.1016/j.patcog.2020.107271


Dense indoor scene modeling from 2D images has been bottlenecked due to the absence of depth information and cluttered occlusions. We present an automatic indoor scene modeling approach using deep features from neural networks. Given a single RGB image, our method simultaneously recovers semantic contents, 3D geometry and object relationship by reasoning indoor environment context. Particularly, we design a shallow-to-deep architecture on the basis of convolutional networks for semantic scene understanding and modeling. It involves multi-level convolutional networks to parse indoor semantics/geometry into non-relational and relational knowledge. Non-relational knowledge extracted from shallow-end networks (e.g. room layout, object geometry) is fed forward into deeper levels to parse relational semantics (e.g. support relationship). A Relation Network is proposed to infer the support relationship between objects. All the structured semantics and geometry above are assembled to guide a global optimization for 3D scene modeling. Qualitative and quantitative analysis demonstrates the feasibility of our method in understanding and modeling semantics-enriched indoor scenes by evaluating the performance of reconstruction accuracy, computation performance and scene complexity.

Item Type:Article
Uncontrolled Keywords:scene understanding; image based modeling; semantic modeling; relational reasoning
Group:Faculty of Media & Communication
ID Code:33683
Deposited By: Symplectic RT2
Deposited On:09 Mar 2020 16:22
Last Modified:14 Mar 2022 14:20


Downloads per month over past year

More statistics for this item...
Repository Staff Only -