Patch Based Synthesis for Single Depth Image Super-Resolution

By Oisin Mac Aodha, Neill D. F. Campbell, Arun Asokan Nair, Gabriel J. Brostow. 2012

234 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

Multi-View Silhouette and Depth Decomposition for High Resolution 3D Object Representation (2018, background): “This research also extends to depth map super-resolution [22, 28, 11].”
Learning Guided Convolutional Network for Depth Completion (2019, methodology): “In terms of depth super-resolution, some methods exploit a database [21] of paired low-resolution and high-resolution depth image patches or self-similarity searching [22] to generate a high resolution depth image.”
Sparsity Invariant CNNs (2017, background): “Early approaches have leveraged repetitive structures to identify similar patches across different scales in 2D [14, 36] and 3D [22].”
Depth and surface normal estimation from monocular images using regression on deep features and hierarchical CRFs (2015, methodology): “Potential 3: Auto-regression model Here we use the auto-regression model to characterize the local correlation structure in the depth map, which has been used in image colorization [25], depth in-painting [4], and depth image super resolution [26, 27].”
Discrete-Continuous Depth Estimation from a Single Image (2014, methodology): “With the recent popularity of depth sensors, sparse depths have also been used to estimate denser depth maps [2].”
Color-Guided Depth Recovery From RGB-D Data Using an Adaptive Autoregressive Model (2014, methodology): “For example, the MRF optmization in [26] needs 12.5 minutes for 4 × upsampling of a 200 × 200 depth map.”
Deep Depth Completion of a Single RGB-D Image (2018, background): “They have exploited a variety of approaches, including Markov random fields [44, 12, 42, 51, 58], shape-from-shading [23, 71], segmentation [41], and dictionary methods [18, 30, 45, 63].”
DeepLiDAR: Deep Surface Normal Guided Depth Prediction for Outdoor Scene From Sparse LiDAR Data and Single Color Image (2018, background): “One family of approach is depth super-resolution that targets improving the resolution of the depth image [35, 43, 15, 53, 32, 23, 36, 47].”
Real-time shading-based refinement for consumer depth cameras (2014, methodology): “One final set of methods increases the resolution of a single depth image offline using a learned database of local patches [Aodha et al. 2012].”
Depth Map Super-Resolution by Deep Multi-Scale Guidance (2016, background): “especially focused on single-image upsampling as MRF labeling problem [16].”