DeLay: Robust Spatial Layout Estimation for Cluttered Indoor Scenes

Dasgupta, Saumitro; Fang, Kuan; Chen, Kevin; Savarese, Silvio

doi:10.1109/cvpr.2016.73

Cited by 134 publications

(118 citation statements)

References 14 publications

Supporting

Mentioning

111

Contrasting

Order By: Relevance

“…Each channel is trained to predict one specific semantic label. To determine the labelling, we used the definition in [14] to avoid ambiguity of semantic labels: If only two walls are visible, they are labelled as a left wall and right wall; if only a single wall is visible, it is labelled as a front wall.…”

Section: Labelsmentioning

confidence: 99%

“…be converted to a single segmentation map, which is a labeling map that represents the semantic surfaces with different labels. Existing work has relied only on either edge maps [12], [13] or semantic labels [14] for layout estimation, which are all learned by fully convolutional neural networks (FCNs) [15]. Although the semantic information was included for multi-task learning in [12] and [13], they were only used to aid the training of the edge maps, and did not add any benefit after training the networks.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Edge-Semantic Learning Strategy for Layout Estimation in Indoor Environment

Zhang

2020

IEEE Trans. Cybern.

View full text Add to dashboard Cite

Visual cognition of the indoor environment can benefit from the spatial layout estimation, which is to represent an indoor scene with a 2D box on a monocular image. In this paper, we propose to fully exploit the edge and semantic information of a room image for layout estimation. More specifically, we present an encoder-decoder network with shared encoder and two separate decoders, which are composed of multiple deconvolution (transposed convolution) layers, to jointly learn the edge maps and semantic labels of a room image. We combine these two network predictions in a scoring function to evaluate the quality of the layouts, which are generated by ray sampling and from a predefined layout pool. Guided by the scoring function, we apply a novel refinement strategy to further optimize the layout hypotheses. Experimental results show that the proposed network can yield accurate estimates of edge maps and semantic labels. By fully utilizing the two different types of labels, the proposed method achieves state-of-the-art layout estimation performance on benchmark datasets.

show abstract

Section: Labelsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Edge-Semantic Learning Strategy for Layout Estimation in Indoor Environment

Zhang

2020

IEEE Trans. Cybern.

View full text Add to dashboard Cite

show abstract

“…A number of researchers have proposed techniques for layout estimation of environments from single images using Manhattan or box world hypotheses, showing excellent results [20,11,12,4]. Our work, which uses two images from different viewpoints and thus enables depth computation via triangulation, could certainly be combined with single-image layout estimation, for example to propagate depth values computed for feature points to whole planar surfaces.…”

Section: Previous Workmentioning

confidence: 99%

“…for k from 1 to 3 do 3: find the two clusters C k i , C k j on same side of L k with minimum Jaccard distance 4:…”

Section: Algorithm 2 Cluster Mergingmentioning

confidence: 99%

Multi-planar Fitting in an Indoor ManhattanWorld

Kim

Manduchi

2017

2017 IEEE Winter Conference on Applications of Computer Vision (WACV)

View full text Add to dashboard Cite

We present an algorithm that finds planar structures in a Manhattan world from two pictures taken from different viewpoints with unknown baseline. The Manhattan world assumption constrains the homographies induced by the visible planes on the image pair, thus enabling robust reconstruction. We extend the T-linkage algorithm for multistructure discovery to account for constrained homographies, and introduce algorithms for sample point selection and orientation-preserving cluster merging. Results are presented on three indoor data set, showing the benefit of the proposed constraints and algorithms.

show abstract

“…Existing solutions for layout estimation mostly rely on hand-crafted features and vanishing lines and they often fail in highly cluttered rooms. More interesting are the approaches of Ren et al [42] and Delay et al [43]. Ren et al, on the one hand, propose a coarse-to-fine strategy for indoor layout estimation.…”

Section: Room Layout Estimation For Location Verificationmentioning

confidence: 99%

Fireground location understanding by semantic linking of visual objects and building information models

Vandecasteele

Merci

Verstockt

2017

Fire Safety Journal

View full text Add to dashboard Cite

This paper presents an outline for improved localization and situational awareness in fire emergency situations based on semantic technology and computer vision techniques. The novelty of our methodology lies in the semantic linking of video object recognition results from visual and thermal cameras with Building Information Models (BIM). The current limitations and possibilities of certain building information streams in the context of fire safety or fire incident management are addressed in this paper. Furthermore, our data management tools match higher-level semantic metadata descriptors of BIM and deep-learning based visual object recognition and classification networks. Based on these matches, estimations can be generated of camera, objects and event positions in the BIM model, transforming it from a static source of information into a rich, dynamic data provider. Previous work has already investigated the possibilities to link BIM and lowcost point sensors for fireground understanding, but these approaches did not take into account the benefits of video analysis and recent developments in semantics and feature learning research. Finally, the strengths of the proposed approach compared to the state-of-the-art is its (semi-)automatic workflow, generic and modular setup and multi-modal strategy, which allows to automatically create situational awareness, to improve localization and to facilitate the overall fire understanding.

show abstract

DeLay: Robust Spatial Layout Estimation for Cluttered Indoor Scenes

Cited by 134 publications

References 14 publications

Edge-Semantic Learning Strategy for Layout Estimation in Indoor Environment

Edge-Semantic Learning Strategy for Layout Estimation in Indoor Environment

Multi-planar Fitting in an Indoor ManhattanWorld

Fireground location understanding by semantic linking of visual objects and building information models

Contact Info

Product

Resources

About