High-accuracy stereo depth maps using structured light

By Daniel Scharstein, Richard Szeliski. 2003

1,870 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology

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 Media Technology

Citations in the corpus

IRS: A Large Synthetic Indoor Robotics Stereo Dataset for Disparity and Surface Normal Estimation (2019, background): “Existing studies [10], [11], [12], [13] proposed stereo datasets collected by real sensing hardware, which contributed a lot to stereo vision research.”
Stereo Matching by Training a Convolutional Neural Network to Compare Image Patches (2015, methodology): “The data sets were published in ﬁve separate works in the years 2001, 2003, 2005, 2006, and 2014 (Scharstein and Szeliski, 2002, 2003; Scharstein and Pal, 2007; Hirschm¨uller and Scharstein, 2007; Scharstein et al., 2014).”
Single Image Dehazing via Multi-scale Convolutional Neural Networks (2016, methodology): “In addition, we generate a validation set of 50 synthesized hazy images using the Middlebury stereo database [34,35,36].”
High-Resolution Stereo Datasets with Subpixel-Accurate Ground Truth (2014, methodology): “Two existing datasets, Cones and Teddy [29], are included on the left for comparison.”
AOD-Net: All-in-One Dehazing Network (2017, methodology): “We also take the 800 full-size synthetic images from the Middlebury stereo database [19, 18, 9] as the TestSet B.”
Learning Conditional Random Fields for Stereo (2007, methodology): “In order to obtain a significant amount of training data for stereo learning approaches, we have created 30 new stereo datasets with ground-truth disparities using an automated version of the structured-lighting technique of [2].”
Large-Scale Data for Multiple-View Stereopsis (2016, methodology): “The reference points, obtained from the structured light scans, are based on binary gray code, which is recommended as being one of the most precise structured light methods (Scharstein and Szeliski 2003; Salvi et al. 2004; Salvi et al. 2010).”
Large Scale Multi-view Stereopsis Evaluation (2014, background): “ommended as being one of the most precise structured light methods [14, 15, 16].”
Dense Semi-rigid Scene Flow Estimation from RGBD Images (2014, methodology): “The Middlebury stereo dataset [15] is commonly used as benchmark to compare scene flow methods [1,8,5,13].”
Fast 3D Scanning with Automatic Motion Compensation (2007, background): “For 3D scanning this ambiguity has often been addressed using a combination of gray-code light and phase-shift [17].”
Multi-Scale Boosted Dehazing Network With Dense Feature Fusion (2020, methodology): “to evaluate the quality of restored images. We note that many existing dehazing methods [50, 9, 43, 41] report their results only on the SOTS indoor images with the models trained on various datasets [33, 55, 54]. Moreover, the GDN method [40] reports the results on the SOTS indoor and outdoor sets,”
Computer Vision for Autonomous Vehicles: Problems, Datasets and State-of-the-Art (2017, background): “Scharstein & Szeliski (2003) and Scharstein et al. (2014) introduced novel datasets to the Middlebury benchmark comprising more complex scenes and including ordinary objects like chairs, tables and plants.”
Multi-view Scene Flow Estimation: A View Centered Variational Approach (2010, methodology): “The Middlebury stereo datasets, Cones, Teddy and V enus [16], were used for generating the data (as in [7]).”
Linear Programming Relaxations and Belief Propagation - An Empirical Study (2006, methodology): “Figure 3 shows the results for a standard stereo benchmark image (the “map” image from the Middlebury stereo benchmark set (Scharstein and Szeliski, 2003)).”
Evaluation of Cost Functions for Stereo Matching (2007, methodology): “For our first set of experiments, we use the standard Middlebury stereo datasets Tsukuba, Venus, Teddy, and Cones [17, 18].”
Spacetime stereo: a unifying framework for depth from triangulation (2005, methodology): “Using an additional real camera has a number of benefits, including improving the robustness of correspondence determination to variations in object reflectance [10], and generating high-quality ground truth stereo test images [29].”
Ieee Transactions on Pattern Analysis and Machine Intelligence 1 Stereo Processing by Semi-global Matching and Mutual Information (background): “Almost all of the currently top-ranked algorithms [2], [3], [7], [9], [11]–[15] on the Tsukuba, Venus, Teddy and Cones data set [18] optimize a global energy function.”
Benchmarking Single-Image Dehazing and Beyond (2017, methodology): “Despite their positive driving effects in the development of dehazing algorithms, those synthetic images are collected from indoor scenes [34], [35], while dehazing is applied to outdoor environments.”