Multimodal templates for real-time detection of texture-less objects in heavily cluttered scenes

Hinterstoißer, Stefan; Hölzer, Stefan M.; Cagniart, Cédric; Ilić, Slobodan; Konolige, Kurt; Navab, Nassir; Lepetit, Vincent

doi:10.1109/iccv.2011.6126326

Cited by 534 publications

(434 citation statements)

References 22 publications

Supporting

Mentioning

433

Contrasting

Order By: Relevance

“…Hinterstoisser et al [9] use simple orientations features densely sampled to build robustness which they call the Dominant Orientation Templates (DOT). Using thousands of such templates for different viewpoints along with multimodal signals (RGB+Depth) they achieve robustness to random clutter.…”

Section: Literature Reviewmentioning

confidence: 99%

“…An important example is in robotics, where the objects of interest can be textureless and often greased or discoloured. However, results from the state-of-the-art algorithm (LineMod) indicate that color is the dominant cue [9] and it outperforms depth cues in the majority of instances that were tested. In LineMod, surface normals were sampled densely over each 2.5D object template and these were used as depth cues for template matching.…”

Section: Introductionmentioning

confidence: 99%

“…Instance recognition in the presence of clutter and occlusion has several important applications, particularly in robotics and augmented reality. Recent methods [9,10,16,6] have begun to exploit the availability of cheap depth sensors to achieve success in instance recognition. Unlike earlier methods that relied on clean laser scanned data [14], these devices have encouraged research into the use of cheap depth data for real-time applications.…”

Section: Introductionmentioning

confidence: 99%

“…Our framework is able to handle real-world instance recognition challenges, i.e, clutter, similar looking distractors and occlusion. Recent algorithms have separately tried to address the problem of clutter [9] and occlusion [16] but fail when these challenges are combined. In comparison we handle all challenges within a single framework.…”

mentioning

confidence: 99%

“…Using occlusion information, computed from the depth data, the forest learns to emphasize the shape features from the visible regions thus making it robust to occlusion. We benchmark our system with the state-of-the-art recognition systems [9,7] in challenging scenes drawn from the largest publicly available dataset. To complement the lack of occlusion tests in this dataset, we introduce our Desk3D dataset and demonstrate that our algorithm outperforms other methods in all settings.…”

mentioning

confidence: 99%

See 4 more Smart Citations

Robust Instance Recognition in Presence of Occlusion and Clutter

Bonde

Badrinarayanan

Cipolla

2014

Computer Vision – ECCV 2014

View full text Add to dashboard Cite

Abstract. We present a robust learning based instance recognition framework from single view point clouds. Our framework is able to handle real-world instance recognition challenges, i.e, clutter, similar looking distractors and occlusion. Recent algorithms have separately tried to address the problem of clutter [9] and occlusion [16] but fail when these challenges are combined. In comparison we handle all challenges within a single framework. Our framework uses a soft label Random Forest [5] to learn discriminative shape features of an object and use them to classify both its location and pose. We propose a novel iterative training scheme for forests which maximizes the margin between classes to improve recognition accuracy, as compared to a conventional training procedure. The learnt forest outperforms template matching, DPM [7] in presence of similar looking distractors. Using occlusion information, computed from the depth data, the forest learns to emphasize the shape features from the visible regions thus making it robust to occlusion. We benchmark our system with the state-of-the-art recognition systems [9,7] in challenging scenes drawn from the largest publicly available dataset. To complement the lack of occlusion tests in this dataset, we introduce our Desk3D dataset and demonstrate that our algorithm outperforms other methods in all settings.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

See 3 more Smart Citations

Robust Instance Recognition in Presence of Occlusion and Clutter

Bonde

Badrinarayanan

Cipolla

2014

Computer Vision – ECCV 2014

View full text Add to dashboard Cite

show abstract

Depth Interpolation Using Tangent Planes on Superpixels of a Color Image

Matsuo

Aoki

2015

Elect Comm in Japan

View full text Add to dashboard Cite

SUMMARY This paper describes a method for accurately interpolating a low‐resolution depth image using a high‐resolution color image. In our method, first, tangent planes on each superpixel are estimated from sparse depth information and dense color information. Then, the neighboring superpixels that have smooth‐connectable tangent planes are connected, and image segmentation to smooth surfaces is achieved. Finally, the low‐resolution depth image is interpolated by using this smooth surface segmentation. In experiments with images from the Middlebury stereo data sets, our method interpolates each image at a high rate, and achieves the lowest error when compared to existing techniques.

show abstract

Human‐robot Teaming for Rescue Missions: Team ViGIR's Approach to the 2013 DARPA Robotics Challenge Trials

Kohlbrecher

Romay

Stumpf

et al. 2014

Journal of Field Robotics

View full text Add to dashboard Cite

Team ViGIR entered the 2013 DARPA Robotics Challenge (DRC) with a focus on developing software to enable an operator to guide a humanoid robot through the series of challenge tasks emulating disaster response scenarios. The overarching philosophy was to make our operators full team members and not just mere supervisors. We designed our operator control station (OCS) to allow multiple operators to request and share information as needed to maintain situational awareness under bandwidth constraints, while directing the robot to perform tasks with most planning and control taking place onboard the robot. Given the limited development time, we leveraged a number of open source libraries in both our onboard software and our OCS design; this included significant use of the robot operating system libraries and toolchain. This paper describes the high level approach, including the OCS design and major onboard components, and it presents our DRC Trials results. The paper concludes with a number of lessons learned that are being applied to the final phase of the competition and are useful for related projects as well. C 2014 Wiley Periodicals, Inc. Kohlbrecher et al.: Human-Robot Teaming for Rescue Missions • 353independence (Huang et al., 2007). The human members of the team function as supervisors who set high-level goals, teammates who assist the robot with perception tasks, and operators who directly change robot parameters to improve performance (Scholtz, 2003); as these roles change dynamically during a set task in our system, we will use the term operator generically. Following Bruemmer et al. (2002), we rarely operate in teleoperation where we directly control a joint value, and we primarily operate in shared mode where the operator specifies tasks or goal points. In shared mode, the robot plans its motions to avoid obstacles and then executes the motion only when given permission. Even when executing a footstep plan in autonomous mode, the operator still has supervisory control of the robot and can command the robot to stop walking at any time and safely revert to a standing posture.Team ViGIR entered the DRC as a "Track B" team competing in the DARPA Virtual Robotics Challenge (VRC). Initially, Team ViGIR was composed of TORC Robotics, 2 the Simulation, Systems Optimization, and Robotics Group at Technische Universität Darmstadt (TUD), 3 and the 3D Interaction Group at Virginia Tech. 4 With only eight months from program kickoff to the first competition, the team focused on providing basic robot capabilities needed for the three tasks in the VRC. A short overview of our VRC approach is available in Kohlbrecher et al. (2013).While the tasks and requirements for the VRC were based on those anticipated in a real scenario, there were important differences: sensor noise was low and more predictable, simple friction models were used, there was no need for calibrating sensors or joint angle offsets for the robot, and the environments were known ahead of time. The dynamic model used for simulating the Atlas robot was ava...

show abstract

Multimodal templates for real-time detection of texture-less objects in heavily cluttered scenes

Cited by 534 publications

References 22 publications

Robust Instance Recognition in Presence of Occlusion and Clutter

Robust Instance Recognition in Presence of Occlusion and Clutter

Depth Interpolation Using Tangent Planes on Superpixels of a Color Image

Human‐robot Teaming for Rescue Missions: Team ViGIR's Approach to the 2013 DARPA Robotics Challenge Trials

Contact Info

Product

Resources

About