Structure from stereo and motion

Richards, Whitman

doi:10.1364/josaa.2.000343

Cited by 121 publications

(91 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, Richards' (1985) scheme predicts that when stereo and kinetic depth are put in conflict, the perceived depth/height ratio will be entirely determined by relative motion whatever the depth/height calculated from binocular disparities. The same result follows from the scheme presented in the Appendix.…”

Section: Discussionmentioning

confidence: 99%

“…As discussed in the Introduction, schemes which combine stereo and motion to solve the distance scaling problem (Richards, 1985; the Appendix to this paper) predict that perceived shape is determined by the shape specified by motion. This is consistent with the results of Expt 1.…”

Section: Methodsmentioning

confidence: 98%

“…Equation (AlO) shows that this produces no change in the value of depth/width or depth/height, simply a uniform scaling of the shape. Consequently, this model of stereo-motion integration, like that of Richards (1985), predicts that the perceived depth/height measured in the ACC task will not be altered by scaling the disparity field to specify another depth of cylinder while leaving the depth specified by motion unchanged.…”

Section: Appendixmentioning

confidence: 97%

“…Nonetheless, given only two pairs of frames from a binocular motion sequence it is still possible to calculate veridical depths (e.g. Richards, 1985). Richards' scheme measures the disparity of individual points in each frame, and uses the ratio of these disparities to extract shape information.…”

Section: Preuious Work On Stereo-motion Combinationmentioning

confidence: 99%

See 3 more Smart Citations

Integration of stereopsis and motion shape cues

Cumming²,

1994

View full text Add to dashboard Cite

A global shape judgement task was used to investigate the combination of stereopsis and kinetic depth. With botb cues present, there were no distortions of shape perception, even under conditions where either cue alone did show such distortions. We suggest that the addition of motion information overcomes the stereo distance scaling problem. However, when incongruent combinations of disparity and motion were used, the results did not match predictions of a number of combination theories. These data could be described by a model which used weighted linear combination afier correctly scaling disparities for viewing distance. When the motion cue was weakened by presenting only two frames of each motion sequence, stereo was weighted more heavily.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 98%

Section: Appendixmentioning

confidence: 97%

Section: Preuious Work On Stereo-motion Combinationmentioning

confidence: 99%

See 2 more Smart Citations

Integration of stereopsis and motion shape cues

Cumming²,

1994

View full text Add to dashboard Cite

show abstract

“…Richards, 1985). Depth from object motion (KDE), for example, does not provide estimates of absolute depth to points of an object, but instead provides estimates which are missing two unknown parameters: the distance to any single point on the object and the direction of object rotation.…”

Section: Cue Promotionmentioning

confidence: 99%

A perturbation analysis of depth perception from combinations of texture and motion cues

1993

View full text Add to dashboard Cite

We examined how depth information from two different cue types (object motion and texture gradient) is integrated into a single estimate in human vision. Two critical assumptions of a recent'model of depth cue combination (termed moa!i@d weak fusion) were tested. The first assumption is that the overall depth estimate is a weighted linear combination of the estimates derived from the individual cues, after initial processing needed to bring them to a common format. The second assumption is that the weight assigned to a cue reflects the apparent reliability of that cue in a particular scene. By this account, the depth combination rule is linear and dynamic, changing in a predictable fashion in response to the particular scene and viewing conditions. A novel procedure was used to measure the weights assigned to the texture and motion cues across experimental conditions. This procedure uses a type of perturbation analysis. The results are consistent with the weighted linear combination rule. In addition, when either cue is corrupted by added noise, the weighted linear combination rule shifts in favor of the uncontaminated cue.

show abstract

Perception and Action Without Veridical Metric Reconstruction: An Affine Approach

Domini

Caudek

2013

Advances in Computer Vision and Pattern Recognition

View full text Add to dashboard Cite

Shape has a long and rich history in vision research. On the computer vision side, shape was the backbone of classical object recognition systems in the 1960s, 1970s, and 1980s. However, the advent of appearance-based recognition in the 1990s drew the spotlight away from shape. While an active shape community continued in the periphery, only recently has shape re-entered the mainstream with a return to contours, shape hierarchies, shape grammars, shape priors, and even 3-D shape inference. On the human vision side, shape research was also affected by paradigm changes. Unlike the computer vision community, psychologists have usually agreed that shape is important, but it has been less clear to them what it is about shape that should be studied: surfaces, invariants, parts, multiple views, learning, simplicity, shape constancy or shape illusions? The growing interest in mathematical formalisms and computational models has begun to provide the long overdue common denominator for these various paradigms.In an effort to foster greater dialog between these two communities of shape researchers, we co-organized a very successful series of four International Workshops on Shape Perception in Human and Computer Vision, in conjunction with ECCV 2008 (http://viper.psych.purdue.edu/workshops/iwsphcv08/), ECVP 2009 (http:// viper.psych.purdue.edu/workshops/iwsphcv09/), ECCV 2010 (http://viper.psych. purdue.edu/workshops/iwsphcv2010/), and VSS 2011 (http://www.visionsciences. org/satellite_shape_perception.htm), two computer vision venues and two human vision venues. The format of each workshop was identical: 12 distinguished invited speakers, 6 from human vision and 6 from computer vision. Each speaker was invited not to present their latest and greatest research, but to reflect more broadly on the issues and challenges they've faced over their careers and the major challenges ahead. Moreover, the speakers were chosen to cover the topic from all sides rather than promote a particular paradigm. The workshops were a great success and received funding from a number of sources.The goal had always been to have the union of the four workshops' authors each submit a chapter to an interdisciplinary collection modeled after the workshops. What follows is a collection that is the realization of that goal, offering 33 chapters by a set of world-class shape researchers from both sides of the aisle. Most of the vii viii Preface authors have worked on the problem of shape perception for many years (decades), and have a unique perspective to offer researchers and students alike on what issues have shaped the field, the trends we've followed, the progress we've made, and the challenges we face. Just like the four workshops, this collection offers a unique, interdisciplinary perspective that is essential for young researchers to understand the broader landscape of the problem so that they can build on a firm foundation. We hope you find the collection as exciting and as useful as we do.There are a number of people and organizations who we'd li...

show abstract

Structure from stereo and motion

Cited by 121 publications

References 14 publications

Integration of stereopsis and motion shape cues

Integration of stereopsis and motion shape cues

A perturbation analysis of depth perception from combinations of texture and motion cues

Perception and Action Without Veridical Metric Reconstruction: An Affine Approach

Contact Info

Product

Resources

About