Shape Anchors for Data-Driven Multi-view Reconstruction

Owens, Andrew; Xiao, Jianxiong; Torralba, Antonio; Freeman, William T.

doi:10.1109/iccv.2013.461

Cited by 16 publications

(11 citation statements)

References 33 publications

(28 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We believe that many new algorithms and applications are enabled by our SUN3D database (e.g. [16]). All source code, labeling tool, and data are publicly available to facilitate further research.…”

Section: Resultsmentioning

confidence: 99%

SUN3D: A Database of Big Spaces Reconstructed Using SfM and Object Labels

Xiao

Owens

Torralba

2013

2013 IEEE International Conference on Computer Vision

Self Cite

706

473

View full text Add to dashboard Cite

Section: Resultsmentioning

confidence: 99%

SUN3D: A Database of Big Spaces Reconstructed Using SfM and Object Labels

Xiao

Owens

Torralba

2013

2013 IEEE International Conference on Computer Vision

Self Cite

706

473

View full text Add to dashboard Cite

“…Liu et al, 2011). In other words, some image patches may be distinctive enough that their latent depth/3D structure can be predicted from their luminance appearance alone (Owens, Xiao, Torralba, & Freeman, 2013). Moreover, depth maps tend to possess simpler, more regular patterns than natural luminance images.…”

Section: Priormentioning

confidence: 99%

Bayesian depth estimation from monocular natural images

Cormack

Bovik

2017

Journal of Vision

View full text Add to dashboard Cite

Estimating an accurate and naturalistic dense depth map from a single monocular photographic image is a difficult problem. Nevertheless, human observers have little difficulty understanding the depth structure implied by photographs. Two-dimensional (2D) images of the real-world environment contain significant statistical information regarding the three-dimensional (3D) structure of the world that the vision system likely exploits to compute perceived depth, monocularly as well as binocularly. Toward understanding how this might be accomplished, we propose a Bayesian model of monocular depth computation that recovers detailed 3D scene structures by extracting reliable, robust, depth-sensitive statistical features from single natural images. These features are derived using well-accepted univariate natural scene statistics (NSS) models and recent bivariate/correlation NSS models that describe the relationships between 2D photographic images and their associated depth maps. This is accomplished by building a dictionary of canonical local depth patterns from which NSS features are extracted as prior information. The dictionary is used to create a multivariate Gaussian mixture (MGM) likelihood model that associates local image features with depth patterns. A simple Bayesian predictor is then used to form spatial depth estimates. The depth results produced by the model, despite its simplicity, correlate well with ground-truth depths measured by a current-generation terrestrial light detection and ranging (LIDAR) scanner. Such a strong form of statistical depth information could be used by the visual system when creating overall estimated depth maps incorporating stereopsis, accommodation, and other conditions. Indeed, even in isolation, the Bayesian predictor delivers depth estimates that are competitive with state-of-the-art "computer vision" methods that utilize highly engineered image features and sophisticated machine learning algorithms.

show abstract

“…In addition to the burden of the hand-crafted features, these approaches require to keep great number of samples to transfer the correct depth to each patch. Moreover, [21] uses this idea to densify an existing depth map, rather than estimate depth from 2D image data only.…”

Section: Related Workmentioning

confidence: 99%

“…An alternative to the models described above is to take a non-parametric, data-intensive approach to depth estimation, notably [1,21]. In these approaches patches from an image are matched to a database of patches each of which is labeled with its correct depth.…”

Section: Related Workmentioning

confidence: 99%

Multi-modal Auto-Encoders as Joint Estimators for Robotics Scene Understanding

Cadena

Dick

Reid

Robotics: Science and Systems XII

View full text Add to dashboard Cite

Abstract-We explore the capabilities of Auto-Encoders to fuse the information available from cameras and depth sensors, and to reconstruct missing data, for scene understanding tasks. In particular we consider three input modalities: RGB images; depth images; and semantic label information. We seek to generate complete scene segmentations and depth maps, given images and partial and/or noisy depth and semantic data. We formulate this objective of reconstructing one or more types of scene data using a Multi-modal stacked Auto-Encoder. We show that suitably designed Multi-modal Auto-Encoders can solve the depth estimation and the semantic segmentation problems simultaneously, in the partial or even complete absence of some of the input modalities. We demonstrate our method using the outdoor dataset KITTI that includes LIDAR and stereo cameras. Our results show that as a means to estimate depth from a single image, our method is comparable to the state-of-the-art, and can run in real time (i.e., less than 40ms per frame). But we also show that our method has a significant advantage over other methods in that it can seamlessly use additional data that may be available, such as a sparse point-cloud and/or incomplete coarse semantic labels.

show abstract

Shape Anchors for Data-Driven Multi-view Reconstruction

Cited by 16 publications

References 33 publications

SUN3D: A Database of Big Spaces Reconstructed Using SfM and Object Labels

SUN3D: A Database of Big Spaces Reconstructed Using SfM and Object Labels

Bayesian depth estimation from monocular natural images

Multi-modal Auto-Encoders as Joint Estimators for Robotics Scene Understanding

Contact Info

Product

Resources

About