Incremental dense semantic stereo fusion for large-scale semantic scene reconstruction

By Vibhav Vineet, Ondrej Miksik, Morten Lidegaard, Matthias Nießner, Stuart Golodetz, Victor Adrian Prisacariu, Olaf Kähler, David William Murray, Shahram Izadi, Patrick Pérez and 2 more. 2015

213 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition, 3D Vision & Pose

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

In 3D Vision & Pose

Citations in the corpus

The Cityscapes Dataset for Semantic Urban Scene Understanding (2016, neutral): “In all cases, we follow the evaluation of the respective dataset to be able to compare to previously reported results [4, 73].”
Past, Present, and Future of Simultaneous Localization and Mapping: Toward the Robust-Perception Age (2016, neutral): “[251] using stereo cameras and a dense map representation.”
Computer Vision for Autonomous Vehicles: Problems, Datasets and State-of-the-Art (2017, background): “CNN-based Methods: Almost all existing algorithms for labeling road scenes are based on machine learning where the parameters of the model are estimated from large annotated datasets.”
Semantic Stereo for Incidental Satellite Images (2018, background): “The two subtasks may be considered separately or sequentially and then fused [26, 27], semantic segmentation may be utilized as a prior for stereo [25, 28], stereo may be utilized as a prior for semantic segmentation [33], or both may be solved together [3, 29].”
SuMa++: Efficient LiDAR-based Semantic SLAM (2019, methodology): “Motivated by the advances of deep learning and Convolutional Neural Networks (CNNs) for scene understanding, there have been many semantic SLAM techniques exploiting this information using cameras [5], [31], cameras + IMU data [4], stereo cameras [10], [15], [18], [33], [38], or RGB-D sensors [3], [”
3DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentation (2018, background): “The analog task can be formulated in 3D, where the goal is to predict semantic labels on a per-voxel level [31,32].”
OpenScene: 3D Scene Understanding with Open Vocabularies (2022, background): “Another line of research performs 3D scene understanding experiments with only 2D ground truth supervisions [16, 27, 38, 47, 53, 57].”
Multi-Label Semantic 3D Reconstruction Using Voxel Blocks (2016, methodology): “Most methods choose to map 2D image segmentation to an existing 3D model [25, 23, 26].”
Tangent Convolutions for Dense Prediction in 3D (2018, background): “A popular line of work exploits graphical models, including conditional random fields [38, 11, 2, 58, 19, 27, 56].”
Visual SLAM for Automated Driving: Exploring the Applications of Deep Learning (2018, methodology): “It has been used in robotics [60, 6, 62, 36], medical applications [14, 71], augmented reality [43], and most prominently automated driving [69, 53, 9, Table 4: Semantic Competition on Cityscapes [15].”
Large-Scale Semantic 3D Reconstruction: An Adaptive Multi-resolution Model for Multi-class Volumetric Labeling (2016, background): “search, which aimed for truly 3-dimensional reconstruction from multiple views, switched to a volumetric representation [17, 2, 25, 36, 39], or in rare cases to meshes [9].”
Meaningful maps with object-oriented semantic mapping (2016, methodology): “[15] proposed an online dense reconstruction method that solves the semantic labelling problem as a densely connected CRF.”
Dense Semantic 3D Reconstruction (2017, methodology): “The method in [39] generates a truncated signed distance field from depth maps using voxel hashing [28] and subsequently assigns semantic labels to the voxels around the surface using an Fig.”
Deep semantic segmentation for automated driving: Taxonomy, roadmap and challenges (2017, methodology): “It has been used in robotics [1][2][3][4], medical applications [5][6], augmented reality [7], and most prominently automated driving [8][9][10][11].”
CNN-SLAM: Real-Time Dense Monocular SLAM with Learned Depth Prediction (2017, background): “Notably, to the best of our knowledge semantic reconstruction has been shown only recently and only based on stereo [28] or RGB-D data [15], i.”
SemanticFusion: Dense 3D semantic mapping with convolutional neural networks (2016, methodology): “[25] produced an incremental 3D reconstruction from stereo pairs and used a Random Forest with a CRF for semantic labelling.”
JSIS3D: Joint Semantic-Instance Segmentation of 3D Point Clouds With Multi-Task Pointwise Networks and Multi-Value Conditional Random Fields (2019, background): “Literature has shown that the data of a 3D scene can be represented by a set of images capturing the scene at different viewpoints [14, 46, 42], in a regular grid of volumes [47, 26, 28], or simply in a 3D point cloud [33, 16, 45, 17, 24].”