Labeling Complete Surfaces in Scene Understanding

Guo, Ruiqi; Hoiem, Derek

doi:10.1007/s11263-014-0776-7

Cited by 8 publications

(4 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition to these transformation invariant features, it is common in building component detection to compute several absolute metrics from the segments. Popular features include the segment's orientation in relation to the gravity or Z-axis, surface area, dimensions and the aspect ratio [4,5,11,[62][63][64][65]. The most frequently used features are summarized in the right column of Table 1.…”

Section: Feature Extractionmentioning

confidence: 99%

Point Cloud vs. Mesh Features for Building Interior Classification

2020

View full text Add to dashboard Cite

Interpreting 3D point cloud data of the interior and exterior of buildings is essential for automated navigation, interaction and 3D reconstruction. However, the direct exploitation of the geometry is challenging due to inherent obstacles such as noise, occlusions, sparsity or variance in the density. Alternatively, 3D mesh geometries derived from point clouds benefit from preprocessing routines that can surmount these obstacles and potentially result in more refined geometry and topology descriptions. In this article, we provide a rigorous comparison of both geometries for scene interpretation. We present an empirical study on the suitability of both geometries for the feature extraction and classification. More specifically, we study the impact for the retrieval of structural building components in a realistic environment which is a major endeavor in Building Information Modeling (BIM) reconstruction. The study runs on segment-based structuration of both geometries and shows that both achieve recognition rates over 75% F1 score when suitable features are used.

show abstract

Section: Feature Extractionmentioning

confidence: 99%

Point Cloud vs. Mesh Features for Building Interior Classification

2020

View full text Add to dashboard Cite

show abstract

“…But errors in layered segmentation can severely degrade the recovery of the occluded region. Learning-based approaches [13,6,44] have posed recovery from occlusion as a 2D semantic segmentation completion task. Ehsani et al [6] propose to complete the silhouette and texture of an occluded object.…”

Section: Related Workmentioning

confidence: 99%

“…Even multi-view approaches [34,12,19] may fail to recover complete shape, since occlusions may block most views of the object. Single-view learning-based methods [13,6,44] have approached seeing beyond occlusion as a 2D semantic segmentation completion task, but complete 3D shape recovery adds the challenges of predicting 3D shape from a 2D image and being robust to the unknown existence and extent of an occluding region.…”

Section: Introductionmentioning

confidence: 99%

Silhouette Guided Point Cloud Reconstruction beyond Occlusion

Zou¹,

Hoiem²

2019

Preprint

Self Cite

View full text Add to dashboard Cite

One major challenge in 3D reconstruction is to infer the complete shape geometry from partial foreground occlusions. In this paper, we propose a method to reconstruct the complete 3D shape of an object from a single RGB image, with robustness to occlusion. Given the image and a silhouette of the visible region, our approach completes the silhouette of the occluded region and then generates a point cloud. We show improvements for reconstruction of nonoccluded and partially occluded objects by providing the predicted complete silhouette as guidance. We also improve state-of-the-art for 3D shape prediction with a 2D reprojection loss from multiple synthetic views and a surface-based smoothing and refinement step. Experiments demonstrate the efficacy of our approach both quantitatively and qualitatively on synthetic and real scene datasets.

show abstract

“…To do so, a dataset of labels exemplars is required. In [14] authors proposed to find the bounding boxes of the objects using object detectors and then regions are classified by combining information from detectors and surface occlusions; they also use RGBdepth to understand the scene.…”

Section: Related Workmentioning

confidence: 99%

Scene Labeling Through Knowledge-Based Rules Employing Constrained Integer Linear Programing

Souly¹,

Shah²

2016

Preprint

View full text Add to dashboard Cite

Scene labeling task is to segment the image into meaningful regions and categorize them into classes of objects which comprised the image. Commonly used methods typically find the local features for each segment and label them using classifiers. Afterward, labeling is smoothed in order to make sure that neighboring regions receive similar labels. However, they ignore expressive and non-local dependencies among regions due to expensive training and inference. In this paper, we propose to use high level knowledge regarding rules in the inference to incorporate dependencies among regions in the image to improve scores of classification. Towards this aim, we extract these rules from data and transform them into constraints for Integer Programming to optimize the structured problem of assigning labels to super-pixels (consequently pixels) of an image. In addition, we propose to use soft-constraints in some scenarios, allowing violating the constraint by imposing a penalty, to make the model more flexible. We assessed our approach on three datasets and obtained promising results.

show abstract

Labeling Complete Surfaces in Scene Understanding

Cited by 8 publications

References 27 publications

Point Cloud vs. Mesh Features for Building Interior Classification

Point Cloud vs. Mesh Features for Building Interior Classification

Silhouette Guided Point Cloud Reconstruction beyond Occlusion

Scene Labeling Through Knowledge-Based Rules Employing Constrained Integer Linear Programing

Contact Info

Product

Resources

About