RGB-(D) scene labeling: Features and algorithms

By Xiaofeng Ren, Liefeng Bo, Dieter Fox. 2012

484 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

How researchers see this paper

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

In Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Semantic Scene Completion from a Single Depth Image (2016, background): “ited to address only part of this problem as shown in Figure 2: RGB-D segmentation approaches consider only visible surfaces without the full 3D shape [6, 26], while shape completion approaches consider only geometry without semantics [3] or a single object out of context [32, 34].”
RefineNet: Multi-path Refinement Networks for High-Resolution Semantic Segmentation (2016, neutral): “[39] RGB-D − 36.”
Bayesian SegNet: Model Uncertainty in Deep Convolutional Encoder-Decoder Architectures for Scene Understanding (2015, background): “CamVid [4] 71.2 45.9 30.7 RGB-D Liu et al. [24] n/a 10.0 n/a Ren et. al [28] n/a 36.3 n/a road scene understanding dataset which has applications for autonomous driving.”
Matterport3D: Learning from RGB-D Data in Indoor Environments (2017, methodology): “These previous RGB-D datasets have been used to train models for several standard scene understanding tasks, including semantic segmentation [27, 33, 13, 40, 7], 3D object detection [31, 24, 14, 36, 37], normal estimation [41, 23, 9, 3], camera relocalization [30, 39], and others [47, 11, 12].”
Joint 2D-3D-Semantic Data for Indoor Scene Understanding (2017, background): “The Computer Vision community has been affected by this change and is experiencing a lot of development in data-driven 2.5D and 3D vision [6, 17, 18, 19, 1].”
Indoor Semantic Segmentation using depth information (2013, methodology): “The results on NYU dataset v1 have been improved by [19] using elaborate kernel descriptors and a post-processing step that employs gPb superpixels MRFs, involving large computation times.”
Deeper Depth Prediction with Fully Convolutional Residual Networks (2016, background): “Moreover, the availability of reasonably accurate depth information is well-known to improve many computer vision tasks with respect to the RGB-only counterpart, for example in reconstruction [23], recognition [26], semantic segmentation [5] or human pose estimation [35].”
Discriminatively Trained Sparse Code Gradients for Contour Detection (2012, unknown): “RGB-D data were also studied in the context of object recognition [13] and scene labeling [27] [22].”
Estimating the 3D Layout of Indoor Scenes and Its Clutter from Depth Sensors (2013, methodology): “Semantic labels: We employ [31] to compute pixel-wise classifiers in terms of 6 furniture classes.”
LinkNet: Exploiting encoder representations for efficient semantic segmentation (2017, background): “A lot of researchers have shifted their focus towards scene understanding because of the surge in tasks like augmented reality and self-driving vehicle, and one of the main step involved in it is pixel-level classification/semantic segmentation [13], [14].”
Semantic Image Segmentation with Contextual Hierarchical Models (2015, methodology): “Beside these appearance features, we also use position and its higher orders (up to second order), which are known to be informative for semantic segmentation [16], [35].”
SegNet: A Deep Convolutional Encoder-Decoder Architecture for Robust Semantic Pixel-Wise Labelling (2015, methodology): “The noisy unaries are then smoothed using a CRF. Improvements were made using a richer feature set including LBP and region segmentation to obtain higher accuracy [28] followed by a CRF.”
Second-order constrained parametric proposals and sequential search-based structured prediction for semantic segmentation in RGB-D images (2015, methodology): “For intensity images, significant progress has been achieved recently through work performed in association with the VOC Semantic Segmentation challenges[11], where high performing methods for boundary detection[36, 31], feature description and nonlinear feature maps[4, 39, 37, 46, 33, 6], image seg”
Aligning 3D models to RGB-D images of cluttered scenes (2015, background): “There have been corresponding works for RGB-D images studying the problems of object detection [4, 13, 17, 20, 21, 22, 24, 34, 35, 36], semantic segmentation [3, 12, 13, 19, 25, 32, 33], and more recently instance segmentation [13, 33].”
Learning Rich Features from RGB-D Images for Object Detection and Segmentation (2014, methodology): “There are multiple ways to approach this problem, like that of doing a bottom-up segmentation and classifying the resulting superpixels [18,30] or modeling contextual relationships among pixels and superpixels [23,33].”
Holistic Scene Understanding for 3D Object Detection with RGBD Cameras (2013, background): “[22, 28, 15] do segmentation based on a variety of carefully designed 2D appearance and 3D geometry features.”
Structure-Preserving Binary Representations for RGB-D Action Recognition (2016, background): “They have been widely applied to many areas such as: human activity recognition [2], robot path planning [3], object detection [4], scene labeling [5], interactive gaming [6], and 3D mapping [7].”