A key requirement for leveraging supervised deep learning methods is the availability of large, labeled datasets. Unfortunately, in the context of RGB-D scene understanding, very little data is available &#x2013; current datasets cover a small range of scene views and have limited semantic annotations. To address this issue, we introduce ScanNet, an RGB-D video dataset containing 2.5M views in 1513 scenes annotated with 3D camera poses, surface reconstructions, and semantic segmentations. To collect this data, we designed an easy-to-use and scalable RGB-D capture system that includes automated surface reconstruction and crowdsourced semantic annotation. We show that using this data helps achieve state-of-the-art performance on several 3D scene understanding tasks, including 3D object classification, semantic voxel labeling, and CAD model retrieval.

/pdf/scannet-richly-annotated-3d-reconstructions-of-indoor-scenes-37ihdazbor.pdf

ScanNet: Richly-Annotated 3D Reconstructions of Indoor Scenes

Simultaneous Localization and Mapping (SLAM)consists in the concurrent construction of a model of the environment (the map), and the estimation of the state of the robot moving within it. The SLAM community has made astonishing progress over the last 30 years, enabling large-scale real-world applications, and witnessing a steady transition of this technology to industry. We survey the current state of SLAM. We start by presenting what is now the de-facto standard formulation for SLAM. We then review related work, covering a broad set of topics including robustness and scalability in long-term mapping, metric and semantic representations for mapping, theoretical performance guarantees, active SLAM and exploration, and other new frontiers. This paper simultaneously serves as a position paper and tutorial to those who are users of SLAM. By looking at the published research with a critical eye, we delineate open challenges and new research issues, that still deserve careful scientific investigation. The paper also contains the authors' take on two questions that often animate discussions during robotics conferences: Do robots need SLAM? and Is SLAM solved?

/pdf/past-present-and-future-of-simultaneous-localization-and-dhljw0akg5.pdf

Past, Present, and Future of Simultaneous Localization And Mapping: Towards the Robust-Perception Age

Although RGB-D sensors have enabled major break-throughs for several vision tasks, such as 3D reconstruction, we have not attained the same level of success in high-level scene understanding. Perhaps one of the main reasons is the lack of a large-scale benchmark with 3D annotations and 3D evaluation metrics. In this paper, we introduce an RGB-D benchmark suite for the goal of advancing the state-of-the-arts in all major scene understanding tasks. Our dataset is captured by four different sensors and contains 10,335 RGB-D images, at a similar scale as PASCAL VOC. The whole dataset is densely annotated and includes 146,617 2D polygons and 64,595 3D bounding boxes with accurate object orientations, as well as a 3D room layout and scene category for each image. This dataset enables us to train data-hungry algorithms for scene-understanding tasks, evaluate them using meaningful 3D metrics, avoid overfitting to a small testing set, and study cross-sensor bias.

/pdf/sun-rgb-d-a-rgb-d-scene-understanding-benchmark-suite-2wsv4bz3ac.pdf

SUN RGB-D: A RGB-D scene understanding benchmark suite

With the invention of the low-cost Microsoft Kinect sensor, high-resolution depth and visual (RGB) sensing has become available for widespread use. The complementary nature of the depth and visual information provided by the Kinect sensor opens up new opportunities to solve fundamental problems in computer vision. This paper presents a comprehensive review of recent Kinect-based computer vision algorithms and applications. The reviewed approaches are classified according to the type of vision problems that can be addressed or enhanced by means of the Kinect sensor. The covered topics include preprocessing, object tracking and recognition, human activity analysis, hand gesture analysis, and indoor 3-D mapping. For each category of methods, we outline their main algorithmic contributions and summarize their advantages/differences compared to their RGB counterparts. Finally, we give an overview of the challenges in this field and future research trends. This paper is expected to serve as a tutorial and source of references for Kinect-based computer vision researchers.

Enhanced Computer Vision With Microsoft Kinect Sensor: A Review

3D shape models are becoming widely available and easier to capture, making available 3D information crucial for progress in object classification. Current state-of-theart methods rely on CNNs to address this problem. Recently, we witness two types of CNNs being developed: CNNs based upon volumetric representations versus CNNs based upon multi-view representations. Empirical results from these two types of CNNs exhibit a large gap, indicating that existing volumetric CNN architectures and approaches are unable to fully exploit the power of 3D representations. In this paper, we aim to improve both volumetric CNNs and multi-view CNNs according to extensive analysis of existing approaches. To this end, we introduce two distinct network architectures of volumetric CNNs. In addition, we examine multi-view CNNs, where we introduce multiresolution filtering in 3D. Overall, we are able to outperform current state-of-the-art methods for both volumetric CNNs and multi-view CNNs. We provide extensive experiments designed to evaluate underlying design choices, thus providing a better understanding of the space of methods available for object classification on 3D data.

/pdf/volumetric-and-multi-view-cnns-for-object-classification-on-4aht612a3h.pdf

Volumetric and Multi-view CNNs for Object Classification on 3D Data

Robust object recognition is a crucial ingredient of many, if not all, real-world robotics applications. This paper leverages recent progress on Convolutional Neural Networks (CNNs) and proposes a novel RGB-D architecture for object recognition. Our architecture is composed of two separate CNN processing streams - one for each modality - which are consecutively combined with a late fusion network. We focus on learning with imperfect sensor data, a typical problem in real-world robotics tasks. For accurate learning, we introduce a multi-stage training methodology and two crucial ingredients for handling depth data with CNNs. The first, an effective encoding of depth information for CNNs that enables learning without the need for large depth datasets. The second, a data augmentation scheme for robust learning with depth images by corrupting them with realistic noise patterns. We present state-of-the-art results on the RGB-D object dataset [15] and show recognition in challenging RGB-D real-world noisy settings.

/pdf/multimodal-deep-learning-for-robust-rgb-d-object-recognition-4yyvywwucv.pdf

Multimodal deep learning for robust RGB-D object recognition

People detection is a key issue for robots and intelligent systems sharing a space with people. Previous works have used cameras and 2D or 3D range finders for this task. In this paper, we present a novel people detection approach for RGB-D data. We take inspiration from the Histogram of Oriented Gradients (HOG) detector to design a robust method to detect people in dense depth data, called Histogram of Oriented Depths (HOD). HOD locally encodes the direction of depth changes and relies on an depth-informed scale-space search that leads to a 3-fold acceleration of the detection process. We then propose Combo-HOD, a RGB-D detector that probabilistically combines HOD and HOG. The experiments include a comprehensive comparison with several alternative detection approaches including visual HOG, several variants of HOD, a geometric person detector for 3D point clouds, and an Haar-based AdaBoost detector. With an equal error rate of 85% in a range up to 8m, the results demonstrate the robustness of HOD and Combo-HOD on a real-world data set collected with a Kinect sensor in a populated indoor environment.

/pdf/people-detection-in-rgb-d-data-24xjt7n323.pdf

People detection in RGB-D data

Developing the perfect SLAM front-end that produces graphs which are free of outliers is generally impossible due to perceptual aliasing. Therefore, optimization back-ends need to be able to deal with outliers resulting from an imperfect front-end. In this paper, we introduce dynamic covariance scaling, a novel approach for effective optimization of constraint networks under the presence of outliers. The key idea is to use a robust function that generalizes classical gating and dynamically rejects outliers without compromising convergence speed. We implemented and thoroughly evaluated our method on publicly available datasets. Compared to recently published state-of-the-art methods, we obtain a substantial speed up without increasing the number of variables in the optimization process. Our method can be easily integrated in almost any SLAM back-end.

/pdf/robust-map-optimization-using-dynamic-covariance-scaling-50885ygb9i.pdf

Robust map optimization using dynamic covariance scaling

Image-based localization is an important problem in robotics and an integral part of visual mapping and navigation systems. An approach to robustly match images to previously recorded ones must be able to cope with seasonal changes especially when it is supposed to work reliably over long periods of time. In this paper, we present a novel approach to visual localization of mobile robots in outdoor environments, which is able to deal with substantial seasonal changes. We formulate image matching as a minimum cost flow problem in a data association graph to effectively exploit sequence information. This allows us to deal with nonmatching image sequences that result from temporal occlusions or from visiting new places. We present extensive experimental evaluations under substantial seasonal changes. Our approach achieves accurate matching across seasons and outperforms existing state-of-the-art methods such as FABMAP2 and SeqSLAM.

/pdf/robust-visual-robot-localization-across-seasons-using-1pkhqxmu6u.pdf

Robust visual robot localization across seasons using network flows

People tracking is a key component for robots that are deployed in populated environments. Previous works have used cameras and 2D and 3D range finders for this task. In this paper, we present a 3D people detection and tracking approach using RGB-D data. We combine a novel multi-cue person detector for RGB-D data with an on-line detector that learns individual target models. The two detectors are integrated into a decisional framework with a multi-hypothesis tracker that controls on-line learning through a track interpretation feedback. For on-line learning, we take a boosting approach using three types of RGB-D features and a confidence maximization search in 3D space. The approach is general in that it neither relies on background learning nor a ground plane assumption. For the evaluation, we collect data in a populated indoor environment using a setup of three Microsoft Kinect sensors with a joint field of view. The results demonstrate reliable 3D tracking of people in RGB-D data and show how the framework is able to avoid drift of the on-line detector and increase the overall tracking performance.

/pdf/people-tracking-in-rgb-d-data-with-on-line-boosted-target-21hacbkwja.pdf

Luciano Spinello

Papers

Multimodal deep learning for robust RGB-D object recognition

People detection in RGB-D data

Robust map optimization using dynamic covariance scaling

Robust visual robot localization across seasons using network flows

People tracking in RGB-D data with on-line boosted target models