Recovering Surface Layout from an Image

By Derek Hoiem, Alexei A. Efros, Martial Hebert. 2007

808 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Environmental Engineering

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Environmental Engineering

Citations in the corpus

SUN database: Large-scale scene recognition from abbey to zoo (2010, methodology): “To make our color and texton histograms more invariant to scene layout, we also build histograms for specific geometric classes as determined by [13].”
Pixel2Mesh: Generating 3D Mesh Models from Single RGB Images (2018, background): “Early works can be traced back to Hoiem et al.[14] and Saxena et al.”
Learning a Predictable and Generative Vector Representation for Objects (2016, background): “Learning-based techniques started gaining traction in the mid2000s [13,29] by framing it as a supervised problem of mapping images of scenes to 2.”
SUN Database: Exploring a Large Collection of Scene Categories (2014, methodology): “To make our color and texton histograms more invariant to scene layout, we also build histograms for specific geometric classes as determined by Hoiem et al. (2007). The geometric classification of a scene is then itself used as a feature, hopefully being invariant to appearance but responsive to la”
Geometric context from a single image (2005, methodology): “To determine which orientation is most likely, we need to use all of the available cues: material, location, texture gradients, shading, vanishing points, etc.”
Approximate Bayesian Image Interpretation using Generative Probabilistic Graphics Programs (2013, background): “Other combinations of top-down knowledge with bottom up processing have been remarkably powerful [9].”
Learning message-passing inference machines for structured prediction (2011, methodology): “The outputs of BPIM-D and [7] are very similar, but we can observe more significant improvements over the M(3)N-F.”
An empirical study of context in object detection (2009, methodology): “The geometric context for an image is computed as a set of seven geometric class (ground, left, right, center, sky, solid, porous) confidence maps as described in [16].”
Recovering free space of indoor scenes from a single image (2012, methodology): “…we show that: (a) our detector works well using image based metrics; (b) our refinement method produces significant improvements in localization in 3D; and (c) if one evaluates using 3D metrics, our method offers major improvements over other single view based scene geometry estimation methods.”
Salient object detection: A survey (2014, background): “he image with different parameters or increasingly grouping superpixels. For example, various levels of spatial support resulting from multiple segmentations are adopted for surface layout prediction [170] and object detection [171]. Interactive segmentation algorithms have also been developed in re”
Probabilistic Joint Image Segmentation and Labeling (2011, background): “While initially inferences were restricted to super-pixel segmentations, recent trends emphasize joint models with capabilities to represent the uncertainty in the segmentation process [2, 4, 5, 6, 7].”
Recovering Occlusion Boundaries from a Single Image (2007, methodology): “To provide further validation, we compare segmentations to two baseline methods: (1) connected components on the most likely surface labels from Hoiem et al. [13]; and (2) a recent version of the NCuts segmentation algorithm [6].”
Perceptual Organization and Recognition of Indoor Scenes from RGB-D Images (2013, background): “Some works have addressed the task of inferring coarse 3D layout of outdoor scenes, exploiting appearance and geometric information [11, 25].”
Towards unified depth and semantic prediction from a single image (2015, background): “MRF-based approaches are common choices in semantic segmentation [1] [36], while they have also been explored in depth prediction [30] [14].”
Efficient Exact Inference for 3D Indoor Scene Understanding (2012, methodology): “Geometric context [12], orientation maps [11] as well as lines in accordance with vanishing points [8] are amongst the most successful cues.”
Displets: Resolving stereo ambiguities using object knowledge (2015, result): “This is in stark contrast to current trends in single image reconstruction [8, 19, 32, 35, 49] where recognition plays a 160 0”
Estimating Spatial Layout of Rooms using Volumetric Reasoning about Objects and Surfaces (2010, methodology): “For comparison, we employ the simple multiple segmentation classifier [8] and the recent approach introduced in [12] as baselines.”
Thinking Inside the Box: Using Appearance Models and Context Based on Room Geometry (2010, background): “Our work also adds to numerous efforts in image-based 3D scene estimation [10–14] and contextual reasoning [15].”
Harmony Potentials (2011, background): “Another way of exploiting global image information is by inferring 3D scene geometry to discover where objects are likely to appear and how big they can be (Hoiem et al. 2007; Hoiem et al. 2008; Munoz et al. 2009).”
Data-Driven 3D Primitives for Single Image Understanding (2013, methodology): “Recently, there has been a renewed push toward more geometric approaches where the appearance of primitives is learned using large amounts of labeled [14] or depth data [23].”