Indoor scene segmentation using a structured light sensor

By Nathan Silberman, Rob Fergus. 2011

550 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

RGB-D datasets using microsoft kinect or similar sensors: a survey (2016, background): “It records video sequences of a great diversity of indoor scenes [79], including a subset of densely labeled video data, raw RGB images, depth images and accelerometer information.”
ScanNet: Richly-Annotated 3D Reconstructions of Indoor Scenes (2017, background): “A large number of RGB-D datasets have been captured and made publicly available for training and benchmarking [56, 34, 50, 65, 79, 83, 74, 4, 58, 81, 15, 55, 1, 68, 30, 51, 21, 48, 43, 92, 80, 61, 72, 93, 36, 16, 35, 57, 40, 29, 70, 52, 45, 95, 75, 9, 33, 85, 71, 32, 3, 10, 78, 2].”
Matterport3D: Learning from RGB-D Data in Indoor Environments (2017, background): “interest in computer vision, graphics, and robotics [10]. Existing work on curation of RGB-D datasets has focused mostly on scans of individual objects [5], standalone rooms [20, 28], views of a room [32, 35], spaces from academic buildings and small apartments [7], and small collections of rooms or”
Indoor Semantic Segmentation using depth information (2013, methodology): “The pioneering work of Silberman et al. [22] was the ﬁrst to deal with the task of semantic full image labeling using depth information.”
N4-Fields: Neural Network Nearest Neighbor Fields for Image Transforms (2014, methodology): “This is demonstrated on three challenging and competitive benchmarks (edge detection on Berkeley Segmentation dataset [4], edge detection on the NYU RGBD dataset[5], retina vessel segmentation on the DRIVE dataset [6]).”
Affordance detection of tool parts from geometric features (2015, background): “depth discontinuities not apparent in color images [31] [28].”
Ieee Transactions on Pattern Analysis and Machine Intelligence 1 3d Reasoning from Blocks to Stability (background): “…that integrating depth with color information improves the performances of many vision tasks, such as segmentation (in [7]), contour detection (in [24]), object recognition (in [25], [26], and [27]), scene labeling (in [28], [29], [30] and [31]), and activity detection (in [32], [33] and [34]).”
Cross-View Semantic Segmentation for Sensing Surroundings (2019, background): “ract this information. However, collecting 3D vision data is costly in time and resources, involving imaging devices such as LiDAR [13], 3D cameras [8, 40], or the extensive scanning of RGB-D cameras [11, 33, 32, 42, 34], as well as the need for a large amount of manual human annotations. In this wo”
Analysis of the Accuracy and Robustness of the Leap Motion Controller (2013, unknown): “Structured light sensors analyze the deformation (warping) of a known pattern onto an unknown surface to determine the three-dimensional shape [6]—representative examples include Microsoft’s Kinect sensor (Kinect For Windows, http://www.microsoft.com/en-us/kinectforwindows) and Asus Xtion Live…”
Indoor Segmentation and Support Inference from RGBD Images (2012, methodology): “They include SIFT features, histograms of surface normals, 2D and 3D bounding box dimensions, color histograms [19] and relative depth [11] 4 .”
3D Scene Understanding by Voxel-CRF (2013, methodology): “We evaluate our framework on two datasets [28, 12].”
ScanNet++: A High-Fidelity Dataset of 3D Indoor Scenes (2023, background): “Early datasets for 3D semantic understanding, such as NYUv2 [37] and SUN RGB-D [38], comprise short RGB-D sequences with low resolution and limited annotations.”
Structured Forests for Fast Edge Detection (2013, methodology): “We assume we are given a set of segmented training images, in which the boundaries between the segments correspond to contours [1, 33].”
Large-Margin Multi-Modal Deep Learning for RGB-D Object Recognition (2015, methodology): “[40] proposed to use 2D and 3D location prior features besides SIFT and spin image.”
Dense 3D semantic mapping of indoor scenes from RGB-D images (2014, methodology): “Semantic segmentation of 2D images recorded from indoor scenes has been shown to be an especially challenging task [6].”
Object detection and classification from large‐scale cluttered indoor scans (2014, methodology): “Object recognition using learning algorithms has been applied to exterior [GF09, GKF09] and indoor environments [BN10, HHF10, TDS10, KAJS11, SF11, KMYG12, LBRF12, NXS12, SXZ ∗ 12].”