Learning invariant object recognition in the visual system with continuous transformations

Stringer, Martin; Perry, Gavin; Rolls, Edmund T.; Proske, H.

doi:10.1007/s00422-005-0030-z

Cited by 68 publications

(99 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Future work involves testing on a larger number of stimuli, and to extend the model to tackle the important challenge of object invariance (cf. [12]). …”

Section: Resultsmentioning

confidence: 99%

Identifying Emotions Using Topographic Conditioning Maps

Pavlou

Casey

2009

Advances in Neuro-Information Processing

View full text Add to dashboard Cite

Abstract. The amygdala is the neural structure that acts as an evaluator of potentially threatening stimuli. We present a biologically plausible model of the visual fear conditioning pathways leading to the amygdala, using a topographic conditioning map (TCM). To evaluate the model, we first use abstract stimuli to understand its ability to form topographic representations, and subsequently to condition on arbitrary stimuli. We then present results on facial emotion recognition using the sub-cortical pathway of the model. Compared to other emotion classification approaches, our model performs well, but does not have the need to pre-specify features. This generic ability to organise visual stimuli is enhanced through conditioning, which also improves classification performance. Our approach demonstrates that a biologically motivated model can be applied to realworld tasks, while allowing us to explore biological hypotheses.

show abstract

“…Future work involves testing on a larger number of stimuli, and to extend the model to tackle the important challenge of object invariance (cf. [12]). …”

Section: Resultsmentioning

confidence: 99%

Identifying Emotions Using Topographic Conditioning Maps

Pavlou

Casey

2009

Advances in Neuro-Information Processing

View full text Add to dashboard Cite

show abstract

“…Interestingly the literature on visual perceptual learning has shown that learning can be surprisingly specific to the precise retinal location of the task stimulus (Fahle 2005). The most influential model of translation invariant object recognition is the so-called trace model (Stringer et al 2006), which assumes that this ability actually depends on learning the activity caused by the same stimulus being shown at many different locations; invariant recognition then emerges at a higher level by learning that these different activations are caused by the same object. Perhaps this is what happens when we learn a tune.…”

Section: Discussionmentioning

confidence: 99%

Relative Pitch Perception and the Detection of Deviant Tone Patterns

Denham

Coath

Háden

et al. 2016

Advances in Experimental Medicine and Biology

View full text Add to dashboard Cite

Most people are able to recognise familiar tunes even when played in a different key. It is assumed that this depends on a general capacity for relative pitch perception; the ability to recognise the pattern of inter-note intervals that characterises the tune. However, when healthy adults are required to detect rare deviant melodic patterns in a sequence of randomly transposed standard patterns they perform close to chance. Musically experienced participants perform better than naïve participants, but even they find the task difficult, despite the fact that musical education includes training in interval recognition.To understand the source of this difficulty we designed an experiment to explore the relative influence of the size of within-pattern intervals and between-pattern transpositions on detecting deviant melodic patterns. We found that task difficulty increases when patterns contain large intervals (5-7 semitones) rather than small intervals (1-3 semitones). While task difficulty increases substantially when transpositions are introduced, the effect of transposition size (large vs small) is weaker. Increasing the range of permissible intervals to be used also makes the task more difficult. Furthermore, providing an initial exact repetition followed by subsequent transpositions does not improve performance. Although musical training correlates with task performance, we find no evidence that violations to musical intervals important in Western music (i.e. the perfect fifth or fourth) are more easily detected. In summary, relative pitch perception does not appear to be conducive to simple explanations based exclusively on invariant physical ratios.

show abstract

“…] Similar results have also been obtained in a hierarchical feedforward network where each layer operates as a competitive network . This network thus captures many of the properties of our hierarchical model of invariant visual object recognition in the ventral visual stream (Elliffe et al 2002;Rolls 1992;Rolls and Deco 2002;Rolls and Milward 2000;Rolls and Stringer 2001, 2006a, 2006bStringer et al 2006;Rolls 2000, 2002;Wallis and Rolls 1997), but incorporates in addition a foveal magnification factor and top^down projections with a dorsal visual stream so that attentional effects can be studied, as shown in figure 7. LGNÐretinal input Figure 7.…”

Section: Change Blindness and Inattentional Blindnessmentioning

confidence: 99%

Top — Down Control of Visual Perception: Attention in Natural Vision

Rolls

2008

Perception

View full text Add to dashboard Cite

IntroductionPre-emptive perception can be used to describe a situation in which a motor command influences perception. An example might be a command for a saccadic eye movement, in which a corollary discharge cancels the changed sensory input produced by the movement of the eyes, contributing to stability of visual perception (Wurtz and Sommer 2004). Here, I consider a somewhat different situation in which prior expectations influence visual perception. These prior expectations might consist of top^down attentional influences on object or spatial processing, or they might consist of prior knowledge of the world which can influence what is seen, as in, for example, the perception of a Necker cube as three-dimensional. A particular focus of this paper is on how attention operates in natural, cluttered, visual scenes, where its influence is much more restricted than when two objects are presented on a blank background. The neurophysiological approach to how attention operates in natural vision leads to hypotheses about the binding problem in perception; about how object representations are interfaced to action systems; and about change blindness. It is shown that even cognitive influences produced by a word descriptor can by top^down processing`pre-empt' affective perception in the olfactory, taste, flavour, somatosensory, and emotional systems, by a top^down modulation of representations in the orbitofrontal cortex. These top^down effects do not pre-empt perception in the sense that they override it. Indeed, it is important that bottom^up inputs dominate the processing to avoid hallucinatory activity. Instead, top^down processes can bias visual perception, in ways for which we now have clear computational-neuroscience models at the level of integrate-and-fire neurons, which show that the top^down effects are most evident when the bottom^up input is weak or ambiguous. The processes described here provide a way for understanding not only top^down effects in attention, but also top^down effects that are Abstract. Top^down perceptual influences can bias (or pre-empt) perception. In natural scenes, the receptive fields of neurons in the inferior temporal visual cortex (IT) shrink to become close to the size of objects. This facilitates the read-out of information from the ventral visual system, because the information is primarily about the object at the fovea. Top^down attentional influences are much less evident in natural scenes than when objects are shown against blank backgrounds, though are still present. It is suggested that the reduced receptive-field size in natural scenes, and the effects of top^down attention contribute to change blindness. The receptive fields of IT neurons in complex scenes, though including the fovea, are frequently asymmetric around the fovea, and it is proposed that this is the solution the IT uses to represent multiple objects and their relative spatial positions in a scene. Networks that implement probabilistic decision-making are described, and it is suggested that, when in perceptua...

show abstract

Learning invariant object recognition in the visual system with continuous transformations

Cited by 68 publications

References 39 publications

Identifying Emotions Using Topographic Conditioning Maps

Identifying Emotions Using Topographic Conditioning Maps

Relative Pitch Perception and the Detection of Deviant Tone Patterns

Top — Down Control of Visual Perception: Attention in Natural Vision

Contact Info

Product

Resources

About