@Book{ 459,
title = {Object Recognition in Man, Monkey, and Machine},
year = {1999},
month = {3},
pages = {217},
abstract = {These interconnected essays on three-dimensional visual object recognition present cutting-edge research by some of the most creative neuroscientific, cognitive, and computational scientists in the field.

Cassandra Moore and Patrick Cavanagh take a classic demonstration, the perception of "two-tone" images, and turn it into a method for understanding the nature of object representations in terms of surfaces and the interaction between bottom-up and top-down processes. Michael J. Tarr and Isabel Gauthier use computer graphics to study whether viewpoint-dependent recognition mechanisms can generalize between exemplars of perceptually defined classes. Melvyn A. Goodale and G. Keith Humphrey use innovative psychophysical techniques to investigate dissociable aspects of visual and spatial processing in brain-injured subjects. D. I. Perrett, M. W. Oram, and E. Ashbridge combine neurophysiological single-cell data from monkeys with computational analyses for a new way of thinking about the mechanisms that mediate viewpoint-dependent object recognition and mental rotation. Shimon Ullman also addresses possible mechanisms to account for viewpoint-dependent behavior, but from the perspective of machine vision. Finally, Philippe G. Schyns synthesizes work from many areas, to provide a coherent account of how stimulus class and recognition task interact.

The contributors bring a wide range of methodologies to bear on the common problem of image-based object recognition.},
web_url = {http://mitpress.mit.edu/catalog/item/default.asp?tid=3885&ttype=2},
publisher = {MIT Press},
address = {Cambridge, MA, USA},
series = {Cognition Special Issues},
state = {published},
ISBN = {978-0-262-70070-2},
author = {Tarr MJ{tarr} and B\"ulthoff HH{hhb}{Department Human Perception, Cognition and Action}}
}

@Article{ 4041,
title = {Recognizing rotated faces and Greebles: What properties drive the face inversion effect?},
journal = {Visual Cognition},
year = {2008},
month = {8},
volume = {16},
number = {6},
pages = {754-784},
abstract = {The fact that faces are strongly affected by picture-plane inversion has often been cited as evidence for face-specific mechanisms. It is unclear, however, whether this "face inversion effect" is driven by properties shared by faces or whether the effect is specific to faces as a category. To address this issue, we compared the recognition of faces and novel Greebles, which were specifically matched to faces along various stimulus dimensions. In two experiments, participants were required to name individual faces or Greebles following training at either single or multiple orientations. We found that performance systematically decreased with increasing misorientation from either the upright (Experiment 1) or nearest trained orientation (Experiment 2). Importantly, the magnitude of this orientation effect was similar for both faces and Greebles. Taken together, these results suggest that the face inversion effect may be a consequence of the visual homogeneity of the stimulus category, regardless of the category
.},
file_url = {/fileadmin/user_upload/files/publications/ashworth_VisualCognition_[0].pdf},
web_url = {http://www.informaworld.com/smpp/ftinterface~content=a780863976~fulltext=713240930},
state = {published},
DOI = {10.1080/13506280701381741},
author = {Ashworth ARS, Vuong QC{qvuong}{Department Human Perception, Cognition and Action}, Rossion B and Tarr MJ{tarr}}
}

@Article{ 2843,
title = {Structural similarity and spatiotemporal noise effects on learning dynamic novel objects},
journal = {Perception},
year = {2006},
month = {3},
volume = {35},
number = {4},
pages = {497-510},
abstract = {The spatiotemporal pattern projected by a moving object is specific to that object, as it
depends on both the object’s shape and its dynamics. Previous research has shown that observers learn to make use of this spatiotemporal signature to recognize dynamic faces
and objects. In two experiments, we assessed the extent to which the structural similarity of the objects and the presence of spatiotemporal noise affect how these signatures are learned and subsequently used in recognition. Observers first learned to identify novel structurally distinctive or structurally similar objects that rotated with a particular motion.
At test, each learned objects moved with its studied motion or with a nonstudied motion. In the nonstudied motion condition we manipulated either dynamic information alone (Experiment 1) or both static and dynamic information (Experiment 2). Across both
experiments we found that changing an object’s learned motion impaired recognition performance when 3D shape was similar or when the visual input was noisy during learning. These results are consistent with the hypothesis that observers use learned spatiotemporal signatures and that such information becomes progressively more important as shape information becomes less reliable.},
file_url = {/fileadmin/user_upload/files/publications/Vuong%20&%20Tarr%202006_2843[0].pdf},
web_url = {http://www.perceptionweb.com/perception/fulltext/p35/p5491.pdf},
state = {published},
DOI = {10.1068/p5491},
author = {Vuong QC{qvuong} and Tarr MJ{tarr}}
}

@Article{ 2599,
title = {The role of surface pigmentation for recognition revealed by contrast reversal in faces and Greebles},
journal = {Vision Research},
year = {2005},
month = {5},
volume = {45},
number = {10},
pages = {1213-1223},
abstract = {Faces are difficult to recognize when viewed as negatives [Galper (1970). Recognition of faces in photographic negative. Psychonomic
Science, 19, 207]. Here we examined the contribution of surface properties to this contrast effect, and whether it is modulated by object category. We tested observers in a matching task using faces or Greebles, presented with or without pigmentation. When
stimulus pairs were shown with mismatched contrast (e.g., positive–negative), there was a decrement in performance. This decrement was larger when the stimuli were shown with pigmentation, and this difference was more pronounced with faces than with Greebles. Overall, contrast reversal disrupts the recognition of both faces and objects to a greater degree in the presence of pigmentation,
suggesting that surface properties are important components of the object representation.},
file_url = {/fileadmin/user_upload/files/publications/Vuong%20Peissig%20et%20al%202005_2599[0].pdf},
web_url = {http://www.sciencedirect.com/science?_ob=MImg&_imagekey=B6T0W-4F29SJ6-3-1&_cdi=4873&_user=29041&_orig=search&_coverDate=05%2F01%2F2005&_sk=999549989&view=c&wchp=dGLzVlz-zSkWW&md5=f1a705c83d65a69c9ecd0e360beb013d&ie=/sdarticle.pdf},
state = {published},
DOI = {10.1016/j.visres.2004.11.015},
author = {Vuong QC{qvuong}{Department Human Perception, Cognition and Action}, Peissig JJ, Harrison MC and Tarr MJ{tarr}}
}

@Article{ 2389,
title = {Rotation direction affects object recognition},
journal = {Vision Research},
year = {2004},
volume = {44},
number = {14},
pages = {1717-1730},
abstract = {What role does dynamic information play in object recognition? To address this question, we probed observers‘ memory for novel objects rotating in depth. Irrespective of object discriminability, performance was affected by an object‘s rotation direction. This effect was obtained despite the same shape information and views being shown for different rotation directions. This direction effect was eliminated when either static images or animations that did not depict globally coherent rotation were used. Overall, these results suggest that dynamic information, that is, the spatiotemporal ordering of object views, provides information independent of shape or view information to a recognition system.},
file_url = {/fileadmin/user_upload/files/publications/Vuong%20&%20Tarr%202004_2389[0].pdf},
state = {published},
author = {Vuong QC{qvuong} and Tarr MJ{tarr}}
}

@Article{ 217,
title = {Why the visual recognition system might encode the effects of illumination},
journal = {Vision Research},
year = {1999},
month = {8},
volume = {38},
number = {15-16},
pages = {2259-2275},
abstract = {A key problem in recognition is that the image of an object depends on the lighting conditions. We investigated whether recognition is sensitive to illumination using 3-D objects that were lit from either the left or right, varying both the shading and the cast shadows. In experiments 1 and 2 participants judged whether two sequentially presented objects were the same regardless of illumination. Experiment 1 used six objects that were easily discriminated and that were rendered with cast shadows. While no cost was found in sensitivity, there was a response time cost over a change in lighting direction. Experiment 2 included six additional objects that were similar to the original six objects making recognition more difficult. The objects were rendered with cast shadows, no shadows, and as a control, white shadows. With normal shadows a change in lighting direction produced costs in both sensitivity and response times. With white shadows there was a much larger cost in sensitivity and a comparable cost in response times. Without cast shadows there was no cost in either measure, but the overall performance was poorer. Experiment 3 used a naming task in which names were assigned to six objects rendered with cast shadows. Participants practised identifying the objects in two viewpoints lit from a single lighting direction. Viewpoint and illumination invariance were then tested over new viewpoints and illuminations. Costs in both sensitivity and response time were found for naming the familiar objects in unfamiliar lighting directions regardless of whether the viewpoint was familiar or unfamiliar. Together these results suggest that illumination effects such as shadow edges: (1) affect visual memory; (2) serve the function of making unambigous the three-dimensional shape; and (3) are modeled with respect to object shape, rather than simply encoded in terms of their effects in the image.},
file_url = {/fileadmin/user_upload/files/publications/why_the_visual_recognition_system_might_encode_the_effects_of_illumination_217[0].pdf},
web_url = {http://www.sciencedirect.com/science?_ob=MImg&_imagekey=B6T0W-3WBFMK9-3-F&_cdi=4873&_user=29041&_pii=S0042698998000418&_origin=&_coverDate=08%2F31%2F1998&_sk=999619984&view=c&wchp=dGLbVlz-zSkWW&md5=d4ebc293868519cfc507935af8fabe7a&ie=/sdarticle.pdf},
state = {published},
DOI = {10.1016/S0042-6989(98)00041-8},
author = {Tarr MJ{tarr}, Kersten D{kersten}{Department Human Perception, Cognition and Action} and B\"ulthoff HH{hhb}{Department Human Perception, Cognition and Action}}
}

@Article{ 181,
title = {What object attributes determine canonical views?},
journal = {Perception},
year = {1999},
month = {5},
volume = {28},
number = {5},
pages = {575-599},
abstract = {We investigated preferred or canonical views for familiar and three-dimensional nonsense objects using computer-graphics psychophysics. We assessed the canonical views for objects by allowing participants to actively rotate realistically shaded three-dimensional models in real-time. Objects were viewed on a Silicon Graphics workstation and manipulated in virtual space with a three-degree-of-freedom input device. In the first experiment, participants adjusted each object to the viewpoint from which they would take a photograph if they planned to use the object to illustrate a brochure. In the second experiment, participants mentally imaged each object on the basis of the name and then adjusted the object to the viewpoint from which they imagined it. In both experiments, there was a large degree of consistency across participants in terms of the preferred view for a given object. Our results provide new insights on the geometrical, experiential, and functional attributes that determine canonical views under ecological conditions.},
file_url = {/fileadmin/user_upload/files/publications/pdf181.pdf},
web_url = {http://www.perceptionweb.com/abstract.cgi?id=p2897},
state = {published},
DOI = {10.1068/p2897},
author = {Blanz V{volker}{Department Human Perception, Cognition and Action}, Tarr MJ{tarr} and B\"ulthoff HH{hhb}{Department Human Perception, Cognition and Action}}
}

@Article{ 259,
title = {Image-based object recognition in man, monkey and machine},
journal = {Cognition},
year = {1998},
month = {7},
volume = {67},
number = {1-2},
pages = {1-20},
abstract = {Theories of visual object recognition must solve the problem of recognizing 3D objects given that perceivers only receive 2D patterns of light on their retinae. Recent findings from human psychophysics, neurophysiology and machine vision provide converging evidence for `image-based' models in which objects are represented as collections of viewpoint-specific local features. This approach is contrasted with `structural-description' models in which objects are represented as configurations of 3D volumes or parts. We then review recent behavioral results that address the biological plausibility of both approaches, as well as some of their computational advantages and limitations. We conclude that, although the image-based approach holds great promise, it has potential pitfalls that may be best overcome by including structural information. Thus, the most viable model of object recognition may be one that incorporates the most appealing aspects of both image-based and structural-description theories.},
file_url = {/fileadmin/user_upload/files/publications/pdf259.pdf},
web_url = {http://www.sciencedirect.com/science?_ob=MImg&_imagekey=B6T24-3TDR775-1-3&_cdi=4908&_user=29041&_pii=S0010027798000262&_origin=&_coverDate=07%2F17%2F1998&_sk=999329998&view=c&wchp=dGLzVlz-zSkWz&md5=c49a8fe35489b1a38503db14759fafba&ie=/sdarticle.pdf},
state = {published},
DOI = {10.1016/S0010-0277(98)00026-2},
author = {Tarr MJ{tarr} and B\"ulthoff HH{hhb}{Department Human Perception, Cognition and Action}}
}

@Article{ 427,
title = {To what extent do unique parts influence recognition across changes in viewpoint?},
journal = {Psychological Science},
year = {1997},
month = {7},
volume = {8},
number = {4},
pages = {282-282},
abstract = {We investigated how varying the number of unique parts within an object influences recognition across changes in viewpoint The stimuli were shaded objects composed of five three-dimensional volumes linked end to end with varying connection angles Of the five volumes, zero, one, three, or five were qualitatively distinct (e g, brick vs cone), the rest being tubes Sequential-matching and naming tasks were used to assess the recognition of these stimuli over rotations in depth Three major results stand out First, regardless of the number of distinct parts, there was increasingly poorer recognition performance with increasing change in viewpoint Second, the impact of viewpoint change for objects with one unique part was less than that for the other objects Third, additional parts beyond a single unique part produced strong viewpoint dependency comparable to that obtained for objects with no distinct parts Thus, visual recognition may be explained by a view-based theory in which viewpoint-specific representations encode both quantitative and qualitative features.},
file_url = {/fileadmin/user_upload/files/publications/pdf427.pdf},
web_url = {http://pss.sagepub.com/content/8/4/282.full.pdf+html},
state = {published},
DOI = {10.1111/j.1467-9280.1997.tb00439.x},
author = {Tarr MJ{tarr}, B\"ulthoff HH{hhb}{Department Human Perception, Cognition and Action}, Zabinski M and Blanz V{volker}{Department Human Perception, Cognition and Action}}
}

@Article{ 670,
title = {Is Human Object Recognition Better Described by Geon Structural Descriptions or by Multiple Views? Comment on Biederman and Gerhardstein (1993)},
journal = {Journal of Experimental Psychology: Human Perception and Performance},
year = {1995},
month = {12},
volume = {21},
number = {6},
pages = {1494-1505},
abstract = {Is human object recognition viewpoint dependent or viewpoint invariant under “everyday” conditions? I. Biederman and P.C. Gerhardstein (1993) argued that viewpoint-invariant mechanisms are used almost exclusively. However, our analysis indicates that (a) their conditions for immediate viewpoint invariance lack the generality to characterize a wide range of recognition phenomena, (b) the extensive body of viewpoint-dependent results cannot be dismissed as processing “by-products” or “experimental artifacts,” and (c) geon structural descriptions cannot coherently account for category recognition, the domain they are intended to explain. The weight of current evidence supports an exemplar-based multiple-views mechanism as an important component of both exemplar-specific and categorical recognition.
*1 This research was supported by U.S. Air Force Office of Scientific Research Contract F49620-91-J-0169 and by the U.S. Office of Naval Research Contract N00014-93-1-0305. Many of the ideas presented in this article were developed during visits by Michael J. Tarr to the Max-Planck-Institut für Biologische Kybernetik, Tübingen, Germany.*2 This article is based on a more detailed version available as Max-Planck Technical Report No. 3, which may be obtained via anonymous ftp to ftp.mpik-tueb.mpg.de as pub/mpi-memos/tr-003.ps.z.*3 We thank Dan Kersten for his insightful comments and thoughtful advice, Shimon Edelman, David Kriegman, Emanuela Bricolo, William Hayward, Laurie Heller, Pepper Williams, and Alice O'Toole for their comments. We also thank William Hayward for initiating the idea expressed in Figure 3 as well as Pierre Jolicoeur for helpful comments.Corresponding Author Contact Information},
file_url = {/fileadmin/user_upload/files/publications/is_human_object_recognition_better_described_by_geon_structural_descriptions_or_by_multiple_views_670[0].pdf},
web_url = {http://www.sciencedirect.com/science?_ob=MImg&_imagekey=B6X08-46RVR0F-18-1&_cdi=7208&_user=29041&_pii=S0096152302006375&_origin=&_coverDate=12%2F31%2F1995&_sk=999789993&view=c&wchp=dGLbVlW-zSkWW&md5=b59174218405285575ecd7c4c92d0b71&ie=/sdarticle.pdf},
state = {published},
author = {Tarr MJ{tarr} and B\"ulthoff HH{hhb}{Department Human Perception, Cognition and Action}}
}

@Article{ 620,
title = {How are three-dimensional objects represented in the brain?},
journal = {Cerebral Cortex},
year = {1995},
month = {5},
volume = {5},
number = {3},
pages = {247-260},
abstract = {In this report we discuss a variety of psychophysical experiments that explore different aspects of the problem of object recognition and
representation in human vision. In all experiments, subjects were presented with realistically rendered images of computer-generated 3D objects, with tight
control over stimulus shape, surface properties, illumination, and viewpoint, as well as subjects‘ prior exposure to the stimulus objects. Contrary to the
predictions of the paradigmatic theory of recognition, which holds that object representations are viewpoint invariant, performance in all experiments was
consistently viewpoint dependent, was only partially aided by binocular stereo and other depth information, was specific to viewpoints that were familiar, and
was systematically disrupted by rotation in depth more than by deforming the 2D images of the stimuli. The emerging concept of multiple-views
representation supported by these results is consistent with recently advanced computational theories of recognition based on view interpolation. Moreover,
in several simulated experiments employing the same stimuli used in experiments with human subjects, models based on multiple-views representations
replicated many of the psychophysical results concerning the observed pattern of human performance.},
file_url = {/fileadmin/user_upload/files/publications/how_are_three_dimensional_objects_represented_in_the_brain_620[0].pdf},
web_url = {http://cercor.oxfordjournals.org/content/5/3/247.long},
state = {published},
DOI = {10.1093/cercor/5.3.247},
author = {B\"ulthoff HH{hhb}{Department Human Perception, Cognition and Action}, Edelman SY{edelman} and Tarr MJ{tarr}}
}

@Inbook{ 2391,
title = {Visual object recognition},
year = {2002},
pages = {287-314},
editor = {Pashler, H. , S. Yantis},
publisher = {John Wiley & Sons, Inc.},
address = {New York, NY},
booktitle = {Steven‘s Handbook of Experimental Psychology (3rd ed.), Vol 1: Sensation and Perception},
state = {published},
author = {Tarr MJ{tarr} and Vuong QC{qvuong}}
}

@Inbook{ 2392,
title = {Neural Representations of Objects},
year = {2001},
pages = {10563-10567},
editor = {Smelser, N. J., P. B. Baltes},
publisher = {Pergamon},
address = {Oxford, UK},
booktitle = {International Encyclopedia of the Social and Behavioral Sciences},
state = {published},
author = {Tarr MJ{tarr} and Vuong QC{qvuong}}
}

@Techreport{ 1507,
title = {What object attributes determine canonical views?},
year = {1996},
month = {11},
number = {42},
abstract = {We investigated preferred or canonical views for familiar and novel three-dimensional objects using computer-graphics psychophysics. We assessed the canonical views for objects by allowing participants to actively rotate realistically shaded three-dimensional models in real-time. Objects were viewed on a Silicon Graphics Workstation and manipulated in virtual space using a three degree-of-freedom input device. In the first experiment, participants adjusted each object to the viewpoint from which they would take a photograph if they planned to use the object to illustrate a brochure. In the second experiment, participants mentally imaged each
object based on the name and then adjusted the object to the viewpoint from which they imagined it. In both experiments, there was a large degree of consistency across participants in terms of the preferred view for a given
object.  Our results provide new insights on the geometrical, experiential, and functional attributes that determine canonical views under ecological conditions.},
file_url = {/fileadmin/user_upload/files/publications/pdf1507.pdf},
state = {published},
author = {Blanz V{volker}{Department Human Perception, Cognition and Action}, Tarr MJ{tarr}, B\"ulthoff HH{hhb}{Department Human Perception, Cognition and Action} and Vetter T{vetter}{Department Human Perception, Cognition and Action}}
}

@Techreport{ 1493,
title = {Scene Recognition Workshop, Tübingen July 3-5, 1996},
year = {1996},
month = {6},
number = {30},
abstract = {In the past few years the question of how humans recognize and analyse natural, complex scenes has received renewed attention, resulting in new techniques and theories. The purpose of this workshop is to bring together researchers interested in these questions from both Europe and the United States to discuss latest results and ideas for the future.},
file_url = {/fileadmin/user_upload/files/publications/pdf1493.pdf},
state = {published},
author = {Tarr MJ{tarr} and B\"ulthoff HH{hhb}{Department Human Perception, Cognition and Action}}
}

@Techreport{ 1486,
title = {To what extent do unique parts influence recognition across changes in viewpoint?},
year = {1995},
month = {9},
number = {22},
abstract = {We investigated how varying the number of unique three-dimensional parts within an object influenced recognition across changes in viewpoint. Stimuli were realistically-shaded images of objects composed of five three-dimensional volumes linked end-to-end. Of the five parts within each object, either zero, one, three, or five were 
qualitatively distinct from other members of the recognition set (e.g., brick versus cone). Non-distinct parts were cylindrical tubes. Independent of the number of distinct parts, the three-dimensional angles between components were different for each object as in Bülthoff and Edelman (1992). In both sequential matching and naming
tasks we compared the impact of depth rotations on recognition performance. Separate between-subject conditions were defined based on the number of distinct parts for each member of the recognition set. The No-Parts condition was run on all subjects and served as a baseline for the other conditions.  For both tasks, three major
results stand out. First, regardless of the number of qualitatively distinct parts there was an effect of viewpoint on recognition performance. Second, the impact of viewpoint change in the One-Part condition was less than that in each of the other conditions.  Third, the addition of parts beyond a single unique part produced strong
viewpoint-dependent recognition performance that was comparable to that obtained for objects with no distinct parts.  Taken together these findings indicate that visual recognition may be accounted for by view-based models in which image-based representations include some qualitatively-defined features.},
note = {This technical report has also been <a href="/main/publication.php?machwas=view_e&edit_lfnr=427">published elsewhere</a>},
state = {published},
author = {Tarr MJ{tarr}, B\"ulthoff HH{hhb}{Department Human Perception, Cognition and Action}, Zabinski M and Blanz V{volker}{Department Human Perception, Cognition and Action}}
}

@Techreport{ 1460,
title = {How are three-dimensional objects represented in the brain?},
year = {1994},
month = {3},
number = {5},
abstract = {We discuss a variety of psychophysical experiments that explore different aspects of the problem of object recognition and representation in human vision.  In all experiments, subjects were presented with realistically rendered images of computer-generated three-dimensional objects, with tight control over stimulus shape, surface properties, illumination, and viewpoint, as well as subjects' prior exposure to the stimulus objects.  Contrary to the predictions of the paradigmatic theory of recognition, which holds that object representations are viewpoint invariant, performance in all experiments was consistently viewpoint dependent, was only partially
aided by binocular stereo and other depth information, was specific to viewpoints that were familiar, and was systematically disrupted by rotation in depth more than by deforming the two-dimensional images of the stimuli. The emerging concept of multiple-view representation
supported by these results is consistent with recently advanced computational theories of recognition based on view
interpolation. Moreover, in several simulated experiments employing the same stimuli used in experiments with human subjects, models based on multiple-views representations replicated many of the psychophysical results concerning the observed pattern of human performance.},
file_url = {/fileadmin/user_upload/files/publications/how_are_three-dimensional_objects_represented_in_the_brain_1460[0].pdf},
state = {published},
author = {B\"ulthoff HH{hhb}{Department Human Perception, Cognition and Action}, Edelman SY{edelman} and Tarr MJ{tarr}}
}

@Techreport{ 682,
title = {How are three-dimensional objects represented in the brain?},
year = {1994},
month = {3},
number = {AIM-1479},
abstract = {We discuss a variety of psychophysical experiments that explore different aspects of the problem of object recognition and representation in human vision.  In all experiments, subjects were presented with realistically rendered images of computer-generated three-dimensional objects, with tight control over stimulus shape, surface properties, illumination, and viewpoint, as well as subjects' prior exposure to the stimulus objects.  Contrary to the predictions of the paradigmatic theory of recognition, which holds that object representations are viewpoint invariant, performance in all experiments was consistently viewpoint dependent, was only partially
aided by binocular stereo and other depth information, was specific to viewpoints that were familiar, and was systematically disrupted by rotation in depth more than by deforming the two-dimensional images of the stimuli. The emerging concept of multiple-view representation
supported by these results is consistent with recently advanced computational theories of recognition based on view
interpolation. Moreover, in several simulated experiments employing the same stimuli used in experiments with human subjects, models based on multiple-views representations replicated many of the psychophysical results concerning the observed pattern of human performance.},
file_url = {/fileadmin/user_upload/files/publications/how_are_three_dimensional_objects_represented_in_the_brain_682[0].pdf},
state = {published},
author = {B\"ulthoff HH{hhb}{Department Human Perception, Cognition and Action}, Edelman SY{edelman} and Tarr MJ{tarr}}
}

@Techreport{ 1457,
title = {Conditions for viewpoint dependence and viewpoint invariance:
What mechanisms are used to recognize an object?},
year = {1993},
month = {12},
number = {3},
abstract = {Is object recognition viewpoint dependent or viewpoint invariant under "everyday" conditions? While Biederman and Gerhardstein (1993) argue that viewpoint-invariant mechanisms are used almost exclusively, an analysis indicates that: 1) their conditions for immediate viewpoint
invariance lack the generality to characterize everyday recognition; 2) viewpoint-dependent effects are {it not} the byproduct of systems other than recognition; 3) empirical evidence supports a prominent role for viewpoint-dependent mechanisms in subordindate-level discriminations; 4) geon structural descriptions provide an inadequate
account of how unfamiliar exemplars of familiar categories are recognized because they are at times too stable and too sensitive with regard to the entry-level. We conclude that exemplar-based multiple-views representations may support both viewpoint-dependent and viewpoint-invariant recognition, with relevant information being applied according to context and task.},
state = {published},
author = {Tarr MJ{tarr} and B\"ulthoff HH{hhb}{Department Human Perception, Cognition and Action}}
}

@Poster{ 4009,
title = {Does contrast reversal affect the recognition of common objects?},
journal = {Journal of Vision},
year = {2006},
month = {6},
volume = {6},
number = {6},
pages = {319},
abstract = {Previous research has shown that faces are difficult to recognize when viewed in reverse contrast (Galper, 1970). In a prior study, we demonstrated that this effect can be found with a set of novel objects (Greebles; Vuong et al. 2005), and that the effect was significantly greater for both faces and novel objects with pigment as opposed to those without pigment. This suggests that surface properties are integral in the representation of both faces and objects. However, it is an open question as to whether surface properties play the same role in the recognition of common objects which are not face-like. In addition, objects and faces shown in normal contrast are on average higher in luminance than those shown in reverse contrast. Here, we normalized all images to be equiluminant to eliminate this possible confound. We tested observers in a same/different sequential-matching task using gray scale images of common objects (cars and birds) as well as faces. Replicating prior results, faces were significantly more difficult to recognize when shown in reverse contrast. Importantly, we also found a contrast effect for the common objects; this effect was the same magnitude as the face contrast effect. These results indicate that performance costs associated with contrast reversal are not due to overall luminance differences between normal and reverse contrast images. Additionally, these data suggest that surfaces properties play a similar role in the recognition of all object types, not just those that are biological or face-like.},
web_url = {http://www.journalofvision.org/content/6/6/319.short},
event_name = {6th Annual Meeting of the Vision Sciences Society (VSS 2006)},
event_place = {Sarasota, FL, USA},
state = {published},
DOI = {10.1167/6.6.319},
author = {Peissig JJ, Vuong QC{qvuong}{Department Human Perception, Cognition and Action}, Vettel JM and Tarr MJ{tarr}}
}

@Poster{ 4634,
title = {Deformation of perceived shape with multiple illumination sources},
journal = {Journal of Vision},
year = {2005},
month = {9},
volume = {5},
number = {8},
pages = {243},
abstract = {The human visual system can easily interpret patterns of shading and attribute them to 3D surfaces. However, perceived shape is rarely veridical. Here we introduce a new effect in which extreme deformations of shape are perceived relative to ground truth. Specifically, convex, specular surfaces illuminated by multiple point-light sources are interpreted in a manner more consistent with one light source illuminating a quite different shape. We hypothesize that the visual system is not able to correctly derive the shape of objects under multiple illuminations due to an overriding single light-source assumption. However, this assumption can be disregarded if there is sufficient evidence against it. For example, other cues to shape such as shadows or stereo disparity may provide information sufficient to support more accurate shape perception, regardless of inferences based on this assumption (although this does not mean that observers now interpret the image as arising from multiple light sources). On the other hand, even the presence of boundary contours may not be sufficient for a “correct” interpretation of the image. Along with psychophysical evidence that observers interpret multiple light-source images as the product of a single source, we developed an image-matching method that produces an image of a shape plus single illuminant that is nearly indistinguishable from the original image with multiple light sources. This method is based on local adjustments of slant in order to minimize the difference between the new image and the target. This method effectively predicts the perceived shape of multiply illuminated convex surfaces. In summary, observers appear to apply relatively simple assumptions in how they derive shape percepts from shading. Moreover, it is possible to capture these assumptions in an image-matching model that accurately predicts observer performance. Thus, our ability to derive accurate models of lighting in a scene may be severely restricted.},
web_url = {http://www.journalofvision.org/5/8/243/},
event_name = {Fifth Annual Meeting of the Vision Sciences Society (VSS 2005)},
event_place = {Sarasota, FL, USA},
state = {published},
digital = {1},
DOI = {10.1167/5.8.243},
author = {Tarr MJ{tarr}, Di Luca M{max} and Zosh WD}
}

@Poster{ 4016,
title = {Learned spatiotemporal signatures are used for object recognition},
year = {2003},
month = {11},
volume = {11},
pages = {9},
web_url = {http://www.opam.net/opam2003/index.htm},
event_name = {11th Annual Workshop on Object Perception, Attention, and Memory (OPAM 2003)},
event_place = {Vancouver, Canada},
state = {published},
author = {Vuong QC{qvuong} and Tarr MJ{tarr}}
}

@Poster{ 4017,
title = {Recognition of familiar objects is impaired by contrast reversal},
year = {2003},
month = {11},
volume = {11},
pages = {6},
web_url = {http://www.opam.net/opam2003/index.htm},
event_name = {11th Annual Workshop on Object Perception, Attention, and Memory (OPAM 2003)},
event_place = {Vancouver, Canada},
state = {published},
author = {Peissig JJ, Vuong QC{qvuong}, Harrison MC and Tarr MJ{tarr}}
}

@Poster{ 4019,
title = {Contrast reversals in faces and objects: The effect of albedo},
journal = {Journal of Vision},
year = {2003},
month = {10},
volume = {3},
number = {9},
pages = {823},
abstract = {Faces are more difficult to recognize than exemplars of other object categories, e.g., chairs, when viewed in reverse contrast (Subramaniam & Biederman, 1997). The visual mechanisms underlying this phenomenon, however, are not well understood. One possibility is that faces and objects rely on separate systems — one sensitive and the other insensitive to contrast reversal. Alternatively, the complex pigmentation patterns characteristic of faces may contribute to this phenomenon (Bruce & Langton, 1994; Liu et al., 1999; Kemp et al., 1996). To date there has not been a direct comparison of the same faces or objects with and without adbedo information (here, defined as the reflectance map of a surface independent of shading/shadows). We tested observers in a same/different sequential-matching task using grayscale images of the same face models with or without albedo information. Observers were instructed to base their judgments on face identity, ignoring any changes in contrast. On trials in which faces were shown in different contrasts (e.g., positive-negative), observers responded more slowly and less accurately as compared to trials in which both faces were shown in the same contrast. This decrement in performance was significantly greater for faces with albedo than for faces without albedo. These results indicate that reported differences between faces and objects across contrast reversal may be attributed to the presence or absence of informative surface information and not to stimulus category. This hypothesis was tested further by comparing the effect of contrast reversal on non-face objects with and without albedo. We conclude that contrast reversal disrupts the recognition of both faces and objects to a greater degree in the presence of informative albedo. That is, another putative dissociation between face and object processing can be accounted for by factors other than the object category per se.},
web_url = {http://www.journalofvision.org/content/3/9/823.abstract},
event_name = {Third Annual Meeting of the Vision Sciences Society (VSS 2003)},
event_place = {Sarasota, FL, USA},
state = {published},
DOI = {10.1167/3.9.823},
author = {Peissig JJ, Vuong QC{qvuong}, Harrison MC and Tarr MJ{tarr}}
}

@Poster{ 4028,
title = {How does object processing change with perceptual expertise?},
journal = {Journal of Vision},
year = {2002},
month = {11},
volume = {2},
number = {7},
pages = {70},
abstract = {To gain a better understanding of how perceptual expertise affects object processing, we trained subjects to be experts at a novel class of homogeneous objects. In contrast to earlier studies, the stimuli, although based on Greebles, were asymmetrical in terms of part arrangement. To assess how object processing changed with expertise, we ran an extensive battery of behavioral tests concurrently with expertise training, including: (1) a same/different task between two sequentially presented images of Greebles where both were either upright or inverted; (2) a same/different task between two simultaneously presented images of Greebles where we varied the presentation time from 50–500 ms; (3) a visual search task where subjects search for a familiar or unfamiliar Greeble among distractors that were either more or less visually similar; (4) an interference task where a Greeble or an object from a non-expert domain (e.g., a car) appeared flanked on both sides by either Greebles or objects from the non-expert class (subjects were told to ignore the flankers and judge whether the target was upright or inverted); and (5) a recognition memory task for familiar versus unfamiliar examples from classes trained at either the individual level (Greebles) or basic level (Fribbles). Based on the assumption that post training configural (holistic) processing is applied automatically to objects from domain of expertise, it was hypothesized that subjects perform differently from their pre-training performance compared to control objects (expertise effects may facilitate or hinder processing). Indeed, this was the case with some tasks showing improved performance and others showing decrements in performance for experts. Such results suggest that perceptual expertise is not a unitary process, but rather recruits a wide variety of mechanisms in the perceptual “tool-kit” that reflect different components of the default processing applied by experts in their domain of expertise.},
web_url = {http://www.journalofvision.org/content/2/7/70.short},
event_name = {Second Annual Meeting of the Vision Sciences Society (VSS 2002)},
event_place = {Sarasota, FL, USA},
state = {published},
DOI = {10.1167/2.7.70},
author = {Kung C-C, Rossion B, Vuong QC{qvuong} and Tarr MJ{tarr}}
}

@Poster{ 4021,
title = {Not all views are created equal: Object identity momentum via dynamic displays},
journal = {Journal of Vision},
year = {2002},
month = {11},
volume = {2},
number = {7},
pages = {41},
abstract = {An object rotating in depth presents a coherent sequence of views but each view is seen only briefly. This scenario raises two questions addressed in the present study: (1) Do observers encode the dynamics of the motion sequence or only the discrete views that comprise the sequence? (2) If motion dynamics are encoded, how do they affect recognition? Observers viewed a movie of an object rotating in depth (study) followed by a static frame showing a single view (test) and judged whether study and test showed the same or different objects. We also manipulated the diagnosticity of the shape information present in the stimulus set and where the test view fell along the rotation trajectory implied by the movie (preceding, early, middle, late, following — all equally spaced). We found that: (1) test frames that followed the trajectory of the motion sequence were better recognized than test frames that preceded the motion sequence; (2) test frames that were actually seen in the sequence were better recognized than following test frames; (3) test frames from the end of the sequence were better recognized than frames from the middle or early part of the sequence. We conclude that observers are biased to encode or remember views that occur near the end of an implied motion trajectory. Moreover, observers use dynamic information to predict unseen views of an object that are consistent with the implied direction of motion. Interestingly, a similar pattern across test conditions was observed for both same and different object trials, suggesting that information encoded about object dynamics is not limited to the particular object being viewed, but instead generalizes to visually-similar objects.},
web_url = {http://w.journalofvision.org/content/2/7/41.short},
event_name = {Second Annual Meeting of the Vision Sciences Society (VSS 2002)},
event_place = {Sarasota, FL, USA},
state = {published},
DOI = {10.1167/2.7.41},
author = {Vuong QC{qvuong} and Tarr MJ{tarr}}
}

@Poster{ 4025,
title = {The role of surface properties in memory for 3D objects.},
journal = {Psychonomics},
year = {2000},
state = {published},
author = {Vuong QC{qvuong} and Tarr MJ{tarr}}
}

@Poster{ 565,
title = {llumination and shadows influence face recognition},
journal = {Investigative Ophthalmology & Visual Science},
year = {1996},
month = {4},
volume = {37},
number = {3},
pages = {176},
event_name = {Annual Meeting of the Association for Research in Vision and Ophthalmology (ARVO 1996)},
event_place = {Fort Lauderdale, FL, USA},
state = {published},
author = {Braje WL, Kersten D{kersten}{Department Human Perception, Cognition and Action}, Tarr MJ{tarr} and Troje NF{niko}{Department Human Perception, Cognition and Action}}
}

@Poster{ 671,
title = {What object attributes determine canonical views?},
journal = {Perception},
year = {1995},
month = {8},
volume = {24},
number = {ECVP Abstract Supplement},
pages = {119-120},
abstract = {We investigated preferred or canonical views for familiar and three-dimensional nonsense objects using computer-graphics psychophysics. We assessed the canonical views for objects by allowing participants to actively rotate realistically shaded three-dimensional models in real-time. Objects were viewed on a Silicon Graphics workstation and manipulated in virtual space with a three-degree-of-freedom input device. In the first experiment, participants adjusted each object to the viewpoint from which they would take a photograph if they planned to use the object to illustrate a brochure. In the second experiment, participants mentally imaged each object on the basis of the name and then adjusted the object to the viewpoint from which they imagined it. In both experiments, there was a large degree of consistency across participants in terms of the preferred view for a given object. Our results provide new insights on the geometrical, experiential, and functional attributes that determine canonical views under ecological conditions.},
event_name = {18th European Conference on Visual Perception},
event_place = {Tübingen, Germany},
state = {published},
author = {Blanz V{volker}{Department Human Perception, Cognition and Action}, Vetter T{vetter}{Department Human Perception, Cognition and Action}, B\"ulthoff HH{hhb}{Department Human Perception, Cognition and Action} and Tarr MJ{tarr}}
}

@Poster{ KerstenTB2012,
title = {Illumination dependency in object recognition},
journal = {Perception},
year = {1994},
month = {9},
volume = {23},
number = {ECVP Abstract Supplement},
pages = {107},
event_name = {Seventeenth European Conference on Visual Perception},
event_place = {Eindhoven, Netherlands},
state = {published},
author = {Kersten D, Tarr M{tarr} and B\"ulthoff HH{hhb}{Department Human Perception, Cognition and Action}}
}

@Conference{ 4031,
title = {Probing the role of motion in object recognition},
year = {2004},
event_name = {Perception for Recognition and Action Workshop},
event_place = {Lunteren, the Netherlands},
state = {published},
author = {Vuong QC{qvuong} and Tarr MJ{tarr}}
}

@Conference{ 673,
title = {To what extent do unique parts influence recognition across changes in viewpoint?},
journal = {Perception},
year = {1995},
month = {8},
volume = {24},
number = {ECVP Abstract Supplement},
pages = {3},
abstract = {We investigated how varying the number of unique three-dimensional parts within an object influenced recognition across changes in viewpoint. Stimuli were realistically-shaded images of objects composed of five three-dimensional volumes linked end-to-end. Of the five parts within each object, either zero, one, three, or five were qualitatively distinct from other members of the recognition set (e.g., brick versus cone). Non-distinct parts were cylindrical tubes. Independent of the number of distinct parts, the three-dimensional angles between components were different for each object as in Bülthoff and Edelman (1992). In both sequential matching and naming tasks we compared the impact of depth rotations on recognition performance. Separate between-subject conditions were defined based on the number of distinct parts for each member of the recognition set. The No-Parts condition was run on all subjects and served as a baseline for the other conditions. For both tasks, three major results stand out. First, regardless of the number of qualitatively distinct parts there was an effect of viewpoint on recognition performance. Second, the impact of viewpoint change in the One-Part condition was less than that in each of the other conditions. Third, the addition of parts beyond a single unique part produced strong viewpoint-dependent recognition performance that was comparable to that obtained for objects with no distinct parts. Taken together these findings indicate that visual recognition may be accounted for by view-based models in which image-based representations include some qualitatively-defined features.},
event_name = {18th European Conference on Visual Perception},
event_place = {Tübingen, Germany},
state = {published},
author = {B\"ulthoff HH{hhb}{Department Human Perception, Cognition and Action}, Tarr MJ{tarr}, Blanz V{volker}{Department Human Perception, Cognition and Action} and Zabinski M}
}