Orientation dependence in the recognition of familiar and novel views of three-dimensional objects

doi:10.1016/0042-6989(92)90102-O

Vision Research

Volume 32, Issue 12, December 1992, Pages 2385-2400

https://doi.org/10.1016/0042-6989(92)90102-O Get rights and content

Abstract

We report four experiments that investigated the representation of novel three-dimensional (3D) objects by the human visual system. In the first experiment, canonical views were demonstrated for novel objects seen equally often from all test viewpoints. The next two experiments showed that the canonical views persisted under repeated testing, and in the presence of a variety of depth cues, including binocular stereo. The fourth experiment probed the ability of subjects to generalize recognition to unfamiliar views of objects previously seen at a limited range of attitudes. Both mono and stereo conditions yielded the same increase in the error rate with misorientation relative to the training attitude. Taken together, these results support the notion that 3D objects are represented by multiple specific views, possibly augmented by partial viewer-centered 3D information.

References (42)

BiedermanI. et al.
Surface versus edge-based determinants of visual recognition
Cognitive Psychology
(1988)
EdelmanS.
Line connectivity algorithms for an asynchronous pyramid computer
Computer Vision, Graphics, and Image Processing
(1987)
RockI. et al.
A case of viewer-centered object perception
Cognitive Psychology
(1987)
RockI. et al.
Can we imagine how objects from other viewpoints?
Cognitive Psychology
(1989)
RoschE. et al.
Basic objects in natural categories
Cognitive Psychology
(1976)
TarrM. et al.
Mental rotation and orientation dependence in shape recognition
Cognitive Psychology
(1989)
UllmanS.
Aligning pictorial descriptions: An approach to object recognition
Cognition
(1989)
BiedermanI.
Recognition by components: A theory of human image understanding
Psychological Review
(1987)
BülthoffH.H.
Shape from X: Stereo, texture, specularity
BülthoffH.H. et al.
Evaluating object recognition theories by computer graphics psychophysics

BülthoffH.H. et al.

Psychophysical support for a 2-D view interpolation theory of object recognition

BülthoffH.H. et al.

Interaction of depth modules: Stereo and shading

Journal of the Optical Society of America

(1988)

BülthoffH.H. et al.

Mapping the generalization space in object recognition

Investigative Ophthalmology and Visual Science (Suppl.)

(1991)

EdelmanS.

Features of recognition

EdelmanS.

A network model of object recognition in human vision

EdelmanS. et al.

Bringing the Grandmother back into the picture: A memory-based view of object recognition

EdelmanS. et al.

A self-organizing multiple-view representation of 3D objects

Biological Cybernetics

(1991)

EdelmanS. et al.

Task and object learning in visual recognition

EdelmanS. et al.

A model of the acquisition of object representations in human 3D visual recognition

FischlerM.A. et al.

Random sample consensus: A paradigm for model fitting with application to image analysis and automated cartography

Communications of the ACM

(1981)

GrimsonW.E.L.

Model-based vision

(1990)

Cited by (348)

Effects of visual cue state and type on the mobile user interface of shopping apps with 360° panoramas
2023, Displays
With the development of modern technology, various shopping apps have emerged, and at the same time, the user interface of shopping apps is constantly being optimized to enhance users’ online shopping experience; in shopping apps, an excellent visual cue design can effectively guide users to quickly complete the process of their visual cognitive processing. The purpose of this study was to explore the usability of visual cue and presentation type in the operation of shopping apps in 360° images and to propose design suggestions for future improvements. A 2 × 3 between-subject design was planned to help explore whether different visual cue states (i.e., dynamic and static cues) and different visual cue presentation types (i.e., dot cue, gesture cue, and cartoon cue) may affect users' task performance and their subjective evaluations. The experiment adopted the purposive sampling method. A total of 60 participants were recruited to take part in the experiment. The experimental data were collected pertinent to task completion time, the system usability scale (SUS), subjective evaluation questionnaires created using a 7-point Likert scale, and semi-structured interviews. The generated results revealed that: (1) Different visual cue states may affect users' performance. (2) Different visual cue presentation types significantly affect users' task performance. Static cues improved users' task performance more than dynamic cues in the visual cue presentation type of dot or cartoon. However, the result was the opposite in the gesture condition. (3) Concerning subjective evaluations of the degree of reasonableness, acceptance, and willingness to use in the condition of the visual cue presentation, the dot or gesture cues are significantly better than the cartoon cue. (4) Among the different visual cue presentation types, participants had a greater preference for the gesture cue. (5) In terms of the degree of reasonableness, acceptance, preference and willingness to use, participants preferred the dynamic cue to the static cue in the condition of the visual cue presentation type of the dot or gesture. However, the opposite result was obtained for the cartoon visual cue presentation type. The findings generated from the research can be a good reference for the development of visual cue design for mobile shopping apps.
Mental rotation in depth as the optical difference of pictures
2023, Methods in Psychology
Mental rotation in depth is a facile interpretation of the Shepard-Metzler effect for perspective pictures. Evidence for that interpretation is given by the linear trends of mean response times over depicted angular disparity, as is found for pairs of pictures of complex solids. Numeric simulation has shown how angular disparity in depth correlates with difference in the profiles of figures in picture pairs. (The profile of a figure is the area contained in its outline against the background, somewhat like its solid angle from a normal view.) Depicted angular disparities in three dimensions are tightly linked to differences in profile across picture pairs, which can be quantified as the autocorrelation coefficient of their difference image. This simple statistic predicts differences in response times for judgments of ‘same’ and ‘different’, and numeric averages over the same statistic mimic the linearity of trend which is characteristic of the Shepard-Metzler effect. The difference in picture profiles is an intervening variable between angular disparity and mean response times which accounts for ‘mental rotation in depth’. An extended demonstration is given of this experimental artifact.
Aesthetic appeal is no coincidence: Preference for generic over specific viewpoints
2022, Vision Research
Generic viewpoints produce images where the visible elements, and their spatial relationships, are largely preserved with small alterations in viewpoint. In contrast, specific viewpoints result from the observer being in a special position relative to the scene, causing spatial alignments between elements that are readily broken with a change in viewpoint. It has been proposed that generic viewpoints are aesthetically more pleasing, due to their lack of suspicious spatial coincidences. We investigated this idea for cartoon foreground and background objects, and found images where foreground and background objects were not suspiciously aligned were judged to be more appealing. A similar, albeit weaker, effect was seen when we digitally manipulated a Monet painting that contained a prominent spatial alignment between a foreground and a background object: the appeal of the painting increased with the magnitude of an introduced spatial offset. A further experiment on an independent observer group found a strong preference for generic viewpoints when tested on a range of cartoon images featuring foregrounds and backgrounds. Our results provide empirical support for the idea that suspicious spatial coincidences between foreground and background objects reduces the aesthetic appeal of images.
Adult-child 3D backward face aging model (3D B-FAM)
2020, Journal of Visual Communication and Image Representation
Face aging has been widely considered in many studies regarding all the potential applications. However, the de-aging known as the rejuvenation or backward modeling has recently received more attention. Previous studies mainly focused on rejuvenating faces from aged adults into young adults using two-dimensional (2D) models. In this work, we propose an extension of a previous 2D adult-child B-FAM into 3D model. This model allows a digital face appearance rejuvenation within a range of [75–3] years old. To evaluate the performances of the proposed approach, first, we proposed two performance evaluation modes, namely: Generic Perception Based and Biometric Verification Mode. Then, the performances have been evaluated over our own 3D database, called Face Time-Machine database constructed using 75 females and 70 males, leading to 500 textured surface meshes. Finally, results show that they are perceptually satisfying and system performance increases by using the faces obtained from our model.
When the whole is equal to the sum of its parts: A new approach to study face and body perception and representation
2019, Vision Research
In this paper, a new approach and a novel method to study face perception is proposed and tested using several qualitative experiments. This method is based on three main tasks: a description task (subjects were asked to freely describe the target stimulus), a free pictorial task (free drawing/painting of what subjects were asked), and a pictorial reproduction task (making a copy of what subjects perceived). These tasks were carried out with children and adults and extended to conditions related to visual arts. The starting points of this work were the canonical perspective and the holistic processes involved in face perception. The aim of this work was to answer the two following basic questions: Are canonical perspective and holistic processes really effective for face perception? Is face perception other than the sum of its parts? The outcomes of the experiments clearly refuted the role of canonical perspective and weaken the holistic approach to face and body perception. The whole human body has been shown instead to appear as if built starting from every single component, therefore body and faces are like wholes, decomposable in a mosaic of juxtaposed independent components reduced to a reference image. In short, the whole is equal to the sum of its parts. Finally, our results also show evidence supporting the introduction of the notion of icon.
The role of perspective in event segmentation
2018, Cognition
People divide their ongoing experience into meaningful events. This process, event segmentation, is strongly associated with visual input: when visual features change, people are more likely to segment. However, the nature of this relationship is unclear. Segmentation could be bound to specific visual features, such as actor posture. Or, it could be based on changes in the activity that are correlated with visual features. This study distinguished between these two possibilities by examining whether segmentation varies across first- and third-person perspectives. In two experiments, observers identified meaningful events in videos of actors performing everyday activities, such as eating breakfast or doing laundry. Each activity was simultaneously recorded from a first-person perspective and a third-person perspective. These videos presented identical activities but differed in their visual features. If segmentation is tightly bound to visual features then observers should identify different events in first- and third-person videos. In addition, the relationship between segmentation and visual features should remain unchanged. Neither prediction was supported. Though participants sometimes identified more events in first-person videos, the events they identified were mostly indistinguishable from those identified for third-person videos. In addition, the relationship between the video’s visual features and segmentation changed across perspectives, further demonstrating a partial dissociation between segmentation and visual input. Event segmentation appears to be robust to large variations in sensory information as long as the content remains the same. Segmentation mechanisms appear to flexibly use sensory information to identify the structure of the underlying activity.

View all citing articles on Scopus

^†: Present address: Department of Applied Mathematics and Computer Science, The Weizmann Institute of Science, Rehovot 76100, Israel.

View full text

Orientation dependence in the recognition of familiar and novel views of three-dimensional objects

Abstract

Cognitive Psychology

Computer Vision, Graphics, and Image Processing

Cognitive Psychology

Cognitive Psychology

Cognitive Psychology

Cognitive Psychology

Cognition

Recognition by components: A theory of human image understanding

Psychological Review

Shape from X: Stereo, texture, specularity

Evaluating object recognition theories by computer graphics psychophysics

Psychophysical support for a 2-D view interpolation theory of object recognition

Interaction of depth modules: Stereo and shading

Journal of the Optical Society of America

Mapping the generalization space in object recognition

Investigative Ophthalmology and Visual Science (Suppl.)

Features of recognition

A network model of object recognition in human vision

Bringing the Grandmother back into the picture: A memory-based view of object recognition

A self-organizing multiple-view representation of 3D objects

Biological Cybernetics

Task and object learning in visual recognition

A model of the acquisition of object representations in human 3D visual recognition

Random sample consensus: A paradigm for model fitting with application to image analysis and automated cartography

Communications of the ACM

Model-based vision