Daniel Huber

Daniel Huber
Carnegie Mellon University | CMU · Robotics Institute

About

90
Publications
36,146
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
6,729
Citations

Publications

Publications (90)
Article
Full-text available
Structural learning, a method to estimate the parameters for discrete energy minimization, has been proven to be effective in solving computer vision problems, especially in 3D scene parsing. As the complexity of the models increases, structural learning algorithms turn to approximate inference to retain tractability. Unfortunately, such methods of...
Conference Paper
Full-text available
Discrete energy minimization is widely-used in computer vision and machine learning for problems such as MAP inference in graphical models. The problem, in general, is notoriously intractable, and finding the global optimal solution is known to be NP-hard. However, is it possible to approximate this problem with a reasonable ratio bound on the solu...
Conference Paper
Many learning-based computer vision algorithms perform poorly when faced with examples that are dissimilar to those on which they were trained. Domain adaptation methods attempt to address this problem, but usually assume that the source domain is specified a priori. We propose a two-step approach for situations where more than one source domain is...
Conference Paper
Applications, such as construction monitoring and planning for renovations, require the accurate recovery of existing conditions of structures. Many types of infrastructure are primarily comprised of arbitrarily-shaped thin structures (e.g., Truss bridges, steel frame buildings under construction, and transmission towers), which existing automatic...
Article
Various object representations have been widely used for many tasks such as object detection, recognition, and tracking. Most of them requires an intensive training process on large database which is collected in advance, and it is hard to add models of a previously unobserved object which is not in the database. In this paper, we investigate how t...
Article
With the emergence of relatively low-cost real-time 3D imaging sensors, new applications for suites of 3D sensors are becoming practical. For example, 3D sensors in an industrial robotic work cell can monitor workers' positions to ensure their safety. This paper introduces a simple-to-use method for extrinsic calibration of multiple 3D sensors obse...
Conference Paper
This paper introduces a method for automatically transforming a point cloud from a laser scanner into a volumetric 3D building model based on the new concept of enclosure reasoning. Rather than simply classifying and modeling building surfaces independently or with pair wise contextual relationships, this work introduces room, floor and building le...
Article
This paper studies the problem of matching images captured from an unmanned ground vehicle (UGV) to those from a satellite or high-flying vehicle. We focus on situations where the UGV navigates in remote areas with few man-made structures. This is a difficult problem due to the drastic change in perspective between the ground and aerial imagery and...
Article
Current GPS-based devices have difficulty localizing in cases where the GPS signal is unavailable or insufficiently accurate. This paper presents an algorithm for localizing a vehicle on an arbitrary road network using vision, road curvature estimates, or a combination of both. The method uses an extension of topometric localization, which is a hyb...
Article
Object representation is useful for many computer vision tasks, such as object detection, recognition, and tracking. Computer vision tasks must handle situations where unknown objects appear and must detect and track some object which is not in the trained database. In such cases, the system must learn or, otherwise derive, descriptions of new obje...
Conference Paper
Localization is a central problem for intelligent vehicles. Visual localization can supplement or replace GPS-based localization approaches in situations where GPS is unavailable or inaccurate. Although visual localization has been demonstrated in a variety of algorithms and systems, the problem of how to best configure such a system remains largel...
Article
Generating three-dimensional (3D) as-is Building Information Models (BIMs), representative of the existing conditions of buildings, from point cloud data collected by laser scanners is becoming common practice. However, generation of such models currently is mostly performed manually, and errors can be introduced during data collection, pre-process...
Article
Full-text available
In the Architecture, Engineering, and Construction (AEC) domain, semantically rich 3D information models are increasingly used throughout a facility's life cycle for diverse applications, such as planning renovations, space usage planning, and managing building maintenance. These models, which are known as building information models (BIMs), are of...
Article
Neya Systems, LLC competed in the CANINE program sponsored by the U.S. Army Tank Automotive Research Development and Engineering Center (TARDEC) which culminated in a competition held at Fort Benning as part of the 2012 Robotics Rodeo. As part of this program, we developed a robot with the capability to learn and recognize the appearance of target...
Conference Paper
We evaluate the performance of a widely used tracking-by-detection and data association multi-target tracking pipeline applied to an activity-rich video dataset. In contrast to traditional work on multi-target pedestrian tracking where people are largely assumed to be upright, we use an activity-rich dataset that includes a wide range of body poses...
Conference Paper
Current manufacturing practices require complete physical separation between people and active industrial robots. These precautions ensure safety, but are inefficient in terms of time and resources, and place limits on the types of tasks that can be performed. In this paper, we present a real-time, sensor-based approach for ensuring the safety of p...
Conference Paper
Full-text available
Autonomous vehicles must be capable of localizing even in GPS denied situations. In this paper, we propose a real-time method to localize a vehicle along a route using visual imagery or range information. Our approach is an implementation of topometric localization, which combines the robustness of topological localization with the geometric accura...
Article
Full-text available
Hasta la fecha, el procesamiento de la información proporcionada por escáners de media distancia en entornos de construcciones civiles y edificios se ha limitado, en la mayoría de los casos, a tareas de registro o generación manual de modelos tridimensionales CAD. En este artículo se da un paso más allá, acometiendo soluciones para problemas de seg...
Conference Paper
Lidar and visual imagery have been broadly utilized in computer vision and mobile robotics applications because these sensors provide complementary information. However, in order to convert data between the local coordinate systems, we must estimate the rigid body transformation between the sensors. In this paper, we propose a robust- weighted extr...
Conference Paper
One of the fundamental requirements of an autonomous vehicle is the ability to determine its location on a map. Frequently, solutions to this localization problem rely on GPS information or use expensive three dimensional (3D) sensors. In this paper, we describe a method for long-term vehicle localization based on visual features alone. Our approac...
Conference Paper
Full-text available
Laser scanners are often used to create 3D models of buildings for civil engineering applications. The current manual process is time-consuming and error-prone. This paper presents a method for using laser scanner data to model predominantly planar surfaces, such as walls, floors, and ceilings, despite the presence of significant amounts of clutter...
Conference Paper
The fusion of stereo and laser range finders (LIDARs) has been proposed as a method to compensate for each individual sensor's deficiencies - stereo output is dense, but noisy for large distances, while LIDAR is more accurate, but sparse. However, stereo usually performs poorly on textureless areas and on scenes containing repetitive structures, an...
Conference Paper
Full-text available
Documenting as-is conditions of buildings using 3D laser scanning and Building Information Modeling (BIM) technology is being adopted as a practice for enhancing effective management of facilities. Many service providers generate as-is BIMs based on laser-scanned data. It is necessary to conduct timely and comprehensive assessments of the quality o...
Conference Paper
Full-text available
The fast and accurate computation of surface normals from a point cloud is a critical step for many 3D robotics and automotive problems, including terrain estimation, mapping, navigation, object segmentation, and object recognition. To obtain the tangent plane to the surface at a point, the traditional approach applies total least squares to its sm...
Article
Full-text available
The task of teleoperating a robot over a wireless video link is known to be very difficult. Teleoperation becomes even more difficult when the robot is surrounded by dense obstacles, or speed requirements are high, or video quality is poor, or wireless links are subject to latency. Due to high-quality lidar data, and improvements in computing and v...
Conference Paper
Full-text available
Evidence grids are a popular representation for fused data from multiple sensors. Previous attempts at back- ground subtraction within evidence grids either do so prior to sensor fusion or do so naively, simply ignoring any cells with a high background occupancy probability. A key weakness of these approaches is that they cannot reason about interi...
Article
Full-text available
In this paper, we present a method for the reconstruction of interiors using a set of panoramic range data in scenes with clutter and occlusion. We specifically deal with the reconstruction of simply-shaped wide areas (such as walls, ceilings and floors) behind furniture and facility pieces in interiors. To date, little attention has been paid to t...
Conference Paper
Full-text available
Three dimensional (3D) imaging sensors, such as laser scanners, are being used to create building information models (BIMs) of the as-is conditions of buildings and other facilities. Quality assurance (QA) needs to be conducted to ensure that the models accurately depict the as-is conditions. We propose a new approach for QA that analyzes patterns...
Article
Full-text available
There is currently no general-purpose, open standard for storing data produced by three dimensional (3D) imaging systems, such as laser scanners. As a result, producers and consumers of such data rely on proprietary or ad-hoc formats to store and exchange data. There is a critical need in the D imaging industry for open standards that promote data...
Article
In many construction and infrastructure management projects, it is important to ensure the flatness of concrete surfaces. Inspectors assess the quality of flat surface construction by checking whether a surface deviates from perfectly flat by more than a specified tolerance. Current flatness assessment methods, such as using a straightedge or shape...
Article
Full-text available
Laser scanner data is increasingly being used for the detailed reverse engineering of buildings. This process is currently primarily manual, but recent research has shown that basic structures, such as walls, ceilings, floors, doorways, and windows, can be detected and modeled automatically. Building on this previous research, we focus on the model...
Article
Full-text available
Laser scanners are increasingly being used to create information rich 3D models of the as-built or as-is conditions of buildings, infrastructure, and other facilities. These "as-built" Building Information Models (BIMs) are created through a time-consuming and error-prone manual process, which is one key barrier to widespread use of as-built BIMs i...
Article
Full-text available
Accurate terrain estimation is critical for au-tonomous offroad navigation. Reconstruction of a 3D surface allows rough and hilly ground to be represented, yielding faster driving and better planning and control. However, data from a 3D sensor samples the terrain unevenly, quickly becoming sparse at longer ranges and containing large voids because...
Article
Building information models (BIMs) are maturing as a new paradigm for storing and exchanging knowledge about a facility. BIMs constructed from a CAD model do not generally capture details of a facility as it was actually built. Laser scanners can be used to capture dense 3D measurements of a facility's as-built condition and the resulting point clo...
Conference Paper
Full-text available
For an autonomous vehicle, detecting and tracking other vehicles is a critical task. Determining the orientation of a detected vehicle is necessary for assessing whether the vehicle is a potential hazard. If a detected vehicle is moving, the orientation can be inferred from its trajectory, but if the vehicle is stationary, the orientation must be d...
Conference Paper
Detecting the boundaries of objects is a key step in separating foreground objects from the background, which is useful for robotics and computer vision applications, such as object detection, recognition, and tracking. We propose a new method for detecting object boundaries using planar laser scanners (LIDARs) and, optionally, co-registered imager...
Conference Paper
Full-text available
For a light rail system, smooth contact between the vehicle and the guide beam is critical for reducing the friction and the vibration of an operating vehicle. Therefore, the shape of guide beams needs to be controlled with mm-level accuracy during the construction. Currently, most methods for detecting shape defects of guide beams, such as experim...
Conference Paper
Full-text available
Laser scanners are rapidly gaining acceptance as a tool for three dimensional (3D) modeling and analysis in the architecture, engineering, and construction (AEC) domain. Since 2001, our cross-disciplinary research team has been developing new methods for analyzing and modeling laser scanner data, with an emphasis on applications in the AEC domain....
Article
Laser scanning is a promising geometric data collection tool for construction, facility, and infrastructure management due to its fast sampling rate (tens of thousands of measurements per second) and millimeter-level accuracy. However, laser scanned data contains inaccurate data points at spatial discontinuities (object edges). These inaccurate poi...
Conference Paper
Scanning laser range sensors (ladars) are frequently used in mobile robotics applications because their ability to accurately measure the environment in 3D makes them well-suited for perception tasks like terrain modeling and obstacle detection. The choice of ladar sensor and the manner in which it is configured and integrated into a robot platform...
Conference Paper
Full-text available
This paper describes a method for creating photorealistic three-dimensional (3D) models of real-world environments in real-time for the purpose of improving and extending the capabilities of vehicle tele-operation. Our approach utilizes the combined data from a laser scanner (for modeling 3D geometry) and a video camera (for modeling surface appear...
Conference Paper
Full-text available
Surface flatness assessment is required for controlling the quality of various products, such as building and mechanical components. During such assessments, inspectors collect data capturing surface shape, and use it to identify flatness defects, which are surface parts deviating from a reference plane by more than the tolerance. Laser scanners ca...
Conference Paper
The task of teleoperating a robot over a wireless video link is known to be very difficult. Teleoperation becomes even more difficult when the robot is surrounded by dense obstacles, or speed requirements are high, or video quality is poor, or wireless links are subject to latency. Due to high quality lidar data, and improvements in computing and v...
Conference Paper
Full-text available
Accurate terrain estimation is critical for autonomous offroad navigation. Reconstruction of a three-dimensional (3D) surface allows rough and hilly ground to be represented, yielding faster driving and better planning and control. However, data from a 3D sensor samples the terrain unevenly, quickly becoming sparse at longer ranges and containing l...
Conference Paper
This paper describes a segmentation method for extracting salient regions in outdoor scenes using both 3-D laser scans and imagery information. Our approach is a bottom- up attentive process without any high-level priors, models, or learning. As a mid-level vision task, it is not only robust against noise and outliers but it also provides valuable...
Article
Full-text available
The assessment of joints with active arthritis is a core component of widely used outcome measures. However, substantial variability exists within and across examiners in assessment of these active joint counts. Swelling and temperature changes, two qualities estimated during active joint counts, are amenable to quantification using noncontact digi...
Conference Paper
Full-text available
Laser scanner measurements are corrupted by noise and artifacts that can undermine the performance of registration, segmentation, surface reconstruction, recognition, and other algorithms operating on the data. While much research has addressed laser scanner noise models, comparatively little is known about other artifacts, such as the mixed pixel...
Article
In recent years, much progress has been made in outdoor autonomous navigation. However, safe navigation is still a daunting challenge in terrain containing vegetation. In this paper, we focus on the segmentation of ladar data into three classes using local three-dimensional point cloud statistics. The classes are: "scatter" to represent porous volu...
Article
Full-text available
We propose a new method for rapid 3D object indexing that combines feature-based methods with coarse alignment-based matching techniques. Our approach achieves a sublinear complexity on the number of models, maintaining at the same time a high degree of performance for real 3D sensed data that is acquired in largely uncontrolled settings. The key c...
Conference Paper
Full-text available
Techniques for three dimensional (3D) imaging and analysis of as-built conditions of buildings are gaining acceptance in the architecture, engineering, and construction (AEC) community. Early detection of defects on construction sites is one domain where these techniques have the potential to revolutionize an industry, since construction defects ca...
Chapter
Full-text available
For several years, our research group has been developing methods for automated modeling of 3D environments. In September, 2002, we were given the opportunity to demonstrate our mapping capability in an underground coal mine. The opportunity arose as a result of the Quecreek mine accident, in which an inaccurate map caused miners to breach an aband...
Article
Defects experienced during construction are costly and preventable. However, inspection programs employed today cannot adequately detect and manage defects that occur on construction sites, as they are based on measurements at specific locations and times, and are not integrated into complete electronic models. Emerging sensing technologies and pro...
Article
For several years, our research group has been developing methods for automated modeling of three-dimensional (3D) environments. In September 2002, we were given the opportunity to demonstrate our mapping capability in an underground coal mine. The opportunity arose as a result of the Quecreek mine accident, in which an inaccurate map caused miners...
Conference Paper
Most research on 3-D object classification and recognition focuses on recognition of objects in 3-D scenes from a small database of known 3-D models. Such an approach does not scale well to large databases of objects and does not generalize well to unknown (but similar) object classification. This paper presents two ideas to address these problems...
Article
Full-text available
Emerging reality capture technologies, such as LADAR and embedded sensing, have potential to increase the efficiency of inspectors by generating detailed data about as-built conditions that can be analyzed in real time and at a later time at an office. The data collected using these systems provide an opportunity to visualize and analyze as-built c...
Article
Full-text available
This paper presents a parts-based method for classifying scenes of 3D objects into a set of pre-determined object classes. Working at the part level, as opposed to the whole object level, enables a more flexible class representation and allows scenes in which the query object is significantly occluded to be classified. In our approach, parts are ex...
Article
This paper proposes a joint feature-based model indexing and geometric constraint based alignment pipeline for efficient and accurate recognition of 3D objects from a large model database. Traditional approaches either first prune the model database using indexing without geometric alignment or directly perform recognition based alignment. The inde...
Conference Paper
Full-text available
Recognition of three dimensional (3D) objects in noisy and cluttered scenes is a challenging problem in 3D computer vision. One approach that has been successful in past research is the regional shape descriptor. In this paper, we introduce two new regional shape descrip- tors: 3D shape contexts and harmonic shape contexts. We evaluate the performa...
Article
Full-text available
Because of the difficulty of interpreting laser data in a meaningful way, safe navigation in vegetated terrain is still a daunting challenge. In this paper, we focus on the segmentation of ladar data using local 3-D point statistics into three classes: clutter to capture grass and tree canopy, linear to capture thin objects like wires or tree branc...