Person:

Dellaert, Frank

Permanent Link

https://hdl.handle.net/1853/71202

Associated Organization(s)

Organizational Unit

School of Interactive Computing

Full item page

Publication Search Results

Now showing 1 - 10 of 18

A Rao-Blackwellized MCMC Algorithm for Recovering Piecewise Planar 3D Models From Multiple View RGBD Images

(Georgia Institute of Technology, 2014-10) Srinivasan, Natesh ; Dellaert, Frank

In this paper, we propose a reconstruction technique that uses 2D regions/superpixels rather than point features. We use pre-segmented RGBD data as input and obtain piecewise planar 3D models of the world. We solve the problem of superpixel labeling within single and multiple views simultaneously by using a Rao-Blackwellized Markov Chain Monte Carlo (MCMC) algorithm. We present our output as a labeled 3D model of the world by integrating out over all possible 3D planes in a fully Bayesian fashion. We present our results on the new SUN3D dataset [?].
Selecting Good Measurements via ℓ₁ Relaxation: A Convex Approach for Robust Estimation Over Graphs

(Georgia Institute of Technology, 2014-09) Carlone, Luca ; Censi, Andrea ; Dellaert, Frank

Pose graph optimization is an elegant and efficient formulation for robot localization and mapping. Experimental evidence suggests that, in real problems, the set of measurements used to estimate robot poses is prone to contain outliers, due to perceptual aliasing and incorrect data association. While several related works deal with the rejection of outliers during pose estimation, the goal of this paper is to propose a grounded strategy for measurements selection, i.e., the output of our approach is a set of “reliable” measurements, rather than pose estimates. Because the classification in inliers /outliers is not observable in general, we pose the problem as finding the maximal subset of the measurements that is internally coherent. In the linear case, we show that the selection of the maximal coherent set can be (conservatively) relaxed to obtain a linear programming problem with ℓ₁ objective. We show that this approach can be extended to (nonlinear) planar pose graph optimization using similar ideas as our previous work on linear approaches to pose graph optimization. We evaluate our method on standard datasets, and we show that it is robust to a large number of outliers and different outlier generation models, while entailing the advantages of linear programming (fast computation, scalability).
Direct Superpixel Labeling for Mobile Robot Navigation Using Learned General Optical Flow Templates

(Georgia Institute of Technology, 2014-09) Roberts, Richard ; Dellaert, Frank

Towards the goal of autonomous obstacle avoidance for mobile robots, we present a method for superpixel labeling using optical flow templates. Optical flow provides a rich source of information that complements image appearance and point clouds in determining traversability. While much past work uses optical flow towards traversability in a heuristic manner, the method we present here instead classifies flow according to several optical flow templates that are specific to the typical environment shape. Our first contribution over prior work in superpixel labeling using optical flow templates is large improvements in accuracy and efficiency by inference directly from spatiotemporal gradients instead of from independently- computed optical flow, and from improved optical flow modeling for obstacles. Our second contribution over the same is extending superpixel labeling methods to arbitrary camera optics without the need to calibrate the camera, by developing and demonstrating a method for learning optical flow templates from unlabeled video. Our experiments demonstrate successful obstacle detection in an outdoor mobile robot dataset.
Mining Structure Fragments for Smart Bundle Adjustment

(Georgia Institute of Technology, 2014-09) Carlone, Luca ; Alcantarilla, Pablo Fernandez ; Chiu, Han-Pang ; Kira, Zsolt ; Dellaert, Frank

Bundle Adjustment (BA) can be seen as an inference process over a factor graph. From this perspective, the Schur complement trick can be interpreted as an ordering choice for elimination. The elimination of a single point in the BA graph induces a factor over the set of cameras observing that point. This factor has a very low information content (a point observation enforces a low-rank constraint on the cameras). In this work we show that, when using conjugate gradient solvers, there is a computational advantage in “grouping” factors corresponding to sets of points (fragments) that are co-visible by the same set of cameras. Intuitively, we collapse many factors with low information content into a single factor that imposes a high-rank constraint among the cameras. We provide a grounded way to group factors: the selection of points that are co-observed by the same camera patterns is a data mining problem, and standard tools for frequent pattern mining can be applied to reveal the structure of BA graphs. We demonstrate the computational advantage of grouping in large BA problems and we show that it enables a consistent reduction of BA time with respect to state-of-the-art solvers (Ceres [1]).
Linear-Time Estimation with Tree Assumed Density Filtering and Low-Rank Approximation

(Georgia Institute of Technology, 2014-09) Ta, Duy-Nguyen ; Dellaert, Frank

We present two fast and memory-efficient approximate estimation methods, targeting obstacle avoidance applications on small robot platforms. Our methods avoid a main bottleneck of traditional filtering techniques, which creates densely correlated cliques of landmarks, leading to expensive time and space complexity. We introduce a novel technique to avoid the dense cliques by sparsifying them into a tree structure and maintain that tree structure efficiently over time. Unlike other edge removal graph sparsification methods, our methods sparsify the landmark cliques by introducing new variables to de-correlate them. The first method projects the current density onto a tree rooted at the same variable at each step. The second method improves upon the first one by carefully choosing a new low-dimensional root variable at each step to replace such that the independence and conditional densities of the landmarks given the trajectory are optimally preserved. Our experiments show a significant improvement in time and space complexity of the methods compared to other standard filtering techniques in worst-case scenarios, with small trade-offs in accuracy due to low-rank approximation errors.
Joint Semantic Segmentation and 3D Reconstruction from Monocular Video

(Georgia Institute of Technology, 2014-09) Kundu, Abhijit ; Li, Yin ; Dellaert, Frank ; Li, Fuxin ; Rehg, James M.

We present an approach for joint inference of 3D scene structure and semantic labeling for monocular video. Starting with monocular image stream, our framework produces a 3D volumetric semantic + occupancy map, which is much more useful than a series of 2D semantic label images or a sparse point cloud produced by traditional semantic segmentation and Structure from Motion(SfM) pipelines respectively. We derive a Conditional Random Field (CRF) model defined in the 3D space, that jointly infers the semantic category and occupancy for each voxel. Such a joint inference in the 3D CRF paves the way for more informed priors and constraints, which is otherwise not possible if solved separately in their traditional frameworks. We make use of class specific semantic cues that constrain the 3D structure in areas, where multiview constraints are weak. Our model comprises of higher order factors, which helps when the depth is unobservable. We also make use of class specific semantic cues to reduce either the degree of such higher order factors, or to approximately model them with unaries if possible. We demonstrate improved 3D structure and temporally consistent semantic segmentation for diffcult, large scale, forward moving monocular image sequence.
iSPCG: Incremental Subgraph-Preconditioned Conjugate Gradient Method for Online SLAM with Many Loop-Closures

(Georgia Institute of Technology, 2014-09) Jian, Yong-Dian ; Dellaert, Frank

We propose a novel method to solve online SLAM problems with many loop-closures on the basis of two state- of-the-art SLAM methods, iSAM and SPCG. We first use iSAM to solve a sparse sub-problem to obtain an approximate solution. When the error grows larger than a threshold or the optimal solution is requested, we use subgraph-preconditioned conjugate gradient method to solve the original problem where the subgraph preconditioner and initial estimate are provided by iSAM. Finally we use the optimal solution from SPCG to regularize iSAM in the next steps. The proposed method is consistent, efficient and can find the optimal solution. We apply this method to solve large simulated and real SLAM problems, and obtain promising results.
SLAM with Object Discovery, Modeling and Mapping

(Georgia Institute of Technology, 2014-09) Choudhary, Siddharth ; Trevor, Alexander J. B. ; Christensen, Henrik I. ; Dellaert, Frank

Object discovery and modeling have been widely studied in the computer vision and robotics communities. SLAM approaches that make use of objects and higher level features have also recently been proposed. Using higher level features provides several benefits: these can be more discriminative, which helps data association, and can serve to inform service robotic tasks that require higher level information, such as object models and poses. We propose an approach for online object discovery and object modeling, and extend a SLAM system to utilize these discovered and modeled objects as landmarks to help localize the robot in an online manner. Such landmarks are particularly useful for detecting loop closures in larger maps. In addition to the map, our system outputs a database of detected object models for use in future SLAM or service robotic tasks. Experimental results are presented to demonstrate the approach’s ability to detect and model objects, as well as to improve SLAM results by detecting loop closures.
Incremental Distributed Robust Inference from Arbitrary Robot Poses via EM and Model Selection

(Georgia Institute of Technology, 2014-07) Indelman, Vadim ; Michael, Nathan ; Dellaert, Frank

We present a novel approach for multi-robot distributed and incremental inference over variables of interest, such as robot trajectories, considering the initial relative poses between the robots and multi-robot data association are both unknown. Assuming robots share with each other informative observations, this inference problem is formulated within an Expectation-Maximization (EM) optimization, performed by each robot separately, alternating between inference over variables of interest and multi-robot data association. To facilitate this process, a common reference frame between the robots should first be established. We show the latter is coupled with determining multi-robot data association, and therefore concurrently infer both using a separate EM optimization. This optimization is performed by each robot starting from several promising initial solutions, converging to locally-optimal hypotheses regarding data association and reference frame transformation. Choosing the best hypothesis in an incremental problem setting is in particular challenging due to high sensitivity to perceptual aliasing and possibly insufficient amount of data. Selecting an incorrect hypothesis introduces outliers and can lead to catastrophic results. To address these challenges we develop a model-selection based approach to choose the most probable hypothesis and use the Chinese restaurant process to disambiguate the hypotheses prior probabilities over time.
An Experimental Study of Robust Distributed Multi-Robot Data Association from Arbitrary Poses

(Georgia Institute of Technology, 2014-06) Nelson, Erik ; Indelman, Vadim ; Michael, Nathan ; Dellaert, Frank

In this work, we experimentally investigate the problem of computing the relative transformation between multiple vehicles from corresponding interrobot observations during autonomous operation in a common unknown environment. Building on our prior work, we consider an EM-based methodology which evaluates sensory observations gathered over vehicle trajectories to establish robust relative pose transformations between robots. We focus on experimentally evaluating the performance of the approach as well as its computational complexity and shared data requirements using multiple autonomous vehicles (aerial robots). We describe an observation subsampling technique which utilizes laser scan autocovariance to reduce the total number of observations shared between robots. Employing this technique reduces run time of the algorithm significantly, while only slightly diminishing the accuracies of computed inter-robot transformations. Finally, we provide discussion on data transfer and the feasibility of implementing the approach on a mesh network.