Abstract-We consider the perceptual challenges inherent in the robotic manipulation of previously unseen socks, with the end goal of manipulation by a household robot for laundry. The task poses challenging problems in modeling the appearance, shape and configuration of these textile items that tend to exhibit high variability in texture, design, and style while being highly articulated objects.At the heart of our approach is a holistic model of shape and appearance that facilitates manipulation of those delicate items-starting even from bunched up instances. We describe novel approaches to two key perceptual problems: (i) Inferring the configuration of the sock, and (ii) determining which socks should be paired together.Robust inference in our model is achieved by strong texture based classifiers that, alone, are powerful enough to solve problems such as inside-out detection. Finally, a reliable prediction of the overall configuration is achieved by combining local cues in a global model that enforces structural consistency.We perform an extensive evaluation of different feature types and classifiers and show strong performance on each subtask of our approach. Finally, we illustrate our approach with an implementation on the Willow Garage PR2-a general purpose robotic platform.
ABSTRACT:This paper proposes an object-space multi-image matching procedure of terrestrial MMS (Mobile Mapping System) image sequences to determine the coordinates of an object point automatically and reliably. This image matching procedure can be applied to find conjugate points of MMS image sequences efficiently. Conventional area-based image matching methods are not reliable to deliver accurate matching results for this application due to image scale variations, viewing angle variations, and object occlusions. In order to deal with these three matching problems, an object space multi-image matching is proposed. A modified NCC (Normalized Cross Correlation) coefficient is proposed to measure the similarity of image patches. A modified multi-window matching procedure will also be introduced to solve the problem of object occlusion. A coarse-to-fine procedure with a combination of object-space multi-image matching and multi-window matching is adopted. The proposed procedure has been implemented for the purpose of matching terrestrial MMS image sequences. The ratio of correct matches of this experiment was about 80%. By providing an approximate conjugate point in an overlapping image manually, most of the incorrect matches could be fixed properly and the ratio of correct matches was improved up to 98%.
A Portable Panoramic Image Mapping System (PPIMS) is proposed for rapid acquisition of three-dimensional spatial information. By considering the convenience of use, cost, weight of equipment, precision, and power supply, the designed PPIMS is equipped with 6 circularly arranged cameras to capture panoramic images and a GPS receiver for positioning. The motivation for this design is to develop a hand-held Mobile Mapping System (MMS) for some difficult accessing areas by vehicle MMS, such as rugged terrains, forest areas, heavily damaged disaster areas, and crowed places etc. This PPIMS is in fact a GPS assisted close-range photogrammetric system. Compared with the traditional close-range photogrammetry, PPIMS can reduce the need of ground control points significantly. Under the condition of knowing the relative geometric relationships of the equipped sensors, the elements of exterior orientation of each captured image can be solved. However, the procedure of a system calibration should be done accurately to determine the relative geometric relationships of multi-cameras and the GPS antenna center, before the PPIMS can be applied for geo-referenced mapping. In this paper, both of one-step and two-step calibration procedures for PPIMS are performed to determine the lever-arm offsets and boresight angles among cameras and GPS. The performance of the one-step and two-step calibration is evaluated through the analysis of the experimental results. The comparison between these two calibration procedures was also conducted. The two-step calibration method outperforms the one-step calibration method in terms of calibration accuracy and operation convenience. We expect that the proposed two-step calibration procedure can also be applied to other platform-based MMSs.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2025 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.