J. Inigo Thomas's research while affiliated with University of Pennsylvania and other places
What is this page?
This page lists the scientific contributions of an author, who either does not have a ResearchGate profile, or has not yet added these contributions to their profile.
It was automatically created by ResearchGate to create a record of this author's body of work. We create such pages to advance our goal of creating and maintaining the most comprehensive scientific repository possible. In doing so, we process publicly available (personal) data relating to the author as a member of the scientific community.
If you're a ResearchGate member, you can follow this page to keep up with this author's work.
If you are this author, and you don't want us to display this page anymore, please let us know.
It was automatically created by ResearchGate to create a record of this author's body of work. We create such pages to advance our goal of creating and maintaining the most comprehensive scientific repository possible. In doing so, we process publicly available (personal) data relating to the author as a member of the scientific community.
If you're a ResearchGate member, you can follow this page to keep up with this author's work.
If you are this author, and you don't want us to display this page anymore, please let us know.
Publications (8)
When a human observer moves, the eye continually fixates on targets in the world. Although fixation is a common process in human vision, its role has not yet been established for computational purposes. The main contribution of this paper is to formalize the retinal flow for a fixating observer. A further contribution - a potentially more practical...
The problem of structure from motion (SFM) is to obtain a D model of a scene using multiple images taken from different camera locations. Traditional SFM algorithms using just two images often produce inaccurate 3D reconstructions, mainly due to incorrect estimation of the camera's motion. More recent algorithms have used multiple (>2) image sequen...
Recovering structure from motion even using information from multiple image frames is difficult, in part because motion error can introduce large, correlated errors in the structure estimate. A method is proposed for recursively recovering structure from motion that can deal with this problem. Encouraging results on real images and synthetic data a...
In order to monitor a robot's position, traditional techniques have used a model of the environment and a 2D view (image) of the scene from the current position of the robot. These techniques require either a fairly accurate 3D scene model or noisy 3D models with a reliable estimate of the model noise. Since it has not been possible either to autom...
In robot navigation a model of the environment needs to be reconstructed for various applications, including path planning, obstacle avoidance and determining where the robot is located. Traditionally, the model was acquired using two images (two--frame Structure from Motion) but the acquired models were unreliable and inaccurate. Recently, researc...
Traditionally, the position of a moving robot is monitored in an
environment by using a 3D model of the scene and a 2D view of the scene
from the current position of the robot. The authors propose automating
the model generation step by using a robust model acquisition algorithm
which also provides an estimate of the error in the acquired model. Th...
This video first summarizes current research at the University of Massachusetts on mobile vehicle navigation using landmark recognition and a partial 3D world model. We then show how landmarks and world models might be automatically acquired and updated over time. A fundamental goal in robot navigation is to determine the "pose" of the robot - that...
Recovering structure from motion even using information from
multiple image frames is difficult, in part because motion error can
introduce large, correlated errors in the structure estimate. They
propose a method for recursively recovering structure from motion that
can deal with this problem. Encouraging results on real images and
synthetic data...
Citations
... The basic idea is constructing a structure estimate by fusing intermediate reconstructions (or subestimates) obtained by processing smaller subsets of the sequence [63]. Moreover, the occlusion problem is handled more easily as a feature should be viewed in only a few frames and not in the entire sequence [49]. ...
... It is well known that with the different types of GPS systems that exist we can obtain positions with errors from 0.02 m to l m. Nevertheless this accuracy cannot be guaranteed all the time in most working environments, where partial satellite occlusion and multipath effects can prevent satisfactory GPS receiver operation910111213. Environmental sensors, such as 3-D scanning laser rangefinder and ultrasonic environmental sensors, are used as well. ...
... Although they use EKF, they also put emphasis on the fact that an equivalent solution can be found using other estimators. Closely related to this example are the work of [Oliensis 1991, Soatto 1993] that are using EKF as a smoothing filter. In the search of better performances, some work investigated the use of optimisation methods based on the minimisation of a nonlinear function [Kumar 1989, Spetsakis 1991, Weng 1993. ...
... Young and Chellappa [75] and Szeliski and Kang [64] analyzed ambiguities in 3D shape and motion estimation along with statistical properties of the estimates. Thomas, Hanson and Oliensis [67] looked at the effects of cross correlation in recursive shape estimation. Daniilidis and Spetsakis [10] analyzed noise sensitivity in as part of the visual navigation problem. ...
![[object Object]](https://i1.rgstatic.net/ii/profile.image/272728041193522-1442034840961_Q64/Allen-Hanson-2.jpg)
... Hence, for the local linearization-based methods, it is not possible to derive suucient conditions for convergence 19]. However, we show that, once motion is estimated, structure is linearly observable in the model (1)-(2), and therefore standard techniques, such as the EKF, can be used eeectively for structure estimation 14, 18, 22]. Therefore the representation described by (1) and (2), though being the very simplest one can imagine, is not the most appropriate for motion estimation. ...
Reference: Motion estimation on the essential manifold
