This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Object Recognition Notes made from chapter 4 of Eysenck and Keane INTRODUCTION Throughout the waking day we are bombarded with information from the visual environment. Mostly we make sense of that information, which usually involves identifying or recognising the objects that surround us. Object recognition typically occurs so effortlessly that it is hard to believe it is actually a rather complex achievement. The complexities of object recognition can be grasped by discussing the processes involved. First, there are usually numerous different overlapping objects in the visual environment, and we must somehow decide where one object ends and the next starts. Second, objects can be recognised accurately over a wide range of viewing distances and orientations. The term ‘CONSTANCY’ refers to the fact that the apparent size and shape of an object do not change despite large variations in the size and shape of the retinal image. Third, we recognise what an object is without any apparent difficulty. Key processes involved in object recognition: · Overlapping: deciding where one object ends and another begins. · Accurate recognition of object over varying distances and orientations. · Allocating diverse visual stimuli to the same category of objects. In spite of the complexities of object recognition, we can generally go beyond simply identifying objects in the visual environment. For example, we can normally describe what an object would look like if viewed form a different angle, and we know its uses and functions. All in all, there is more to object recognition than might be initially supposed. MARR’S COMPUTATIONAL THEORY MARR (1982) put forward a COMPUTATIONAL THEORY of the processes involved in object recognition. He proposed a series of REPRESENTATIONS (i.e. descriptions) providing increasingly detailed information about the visual environment. MARR identified 3 major kinds of representation: · PRIMAL SKETCH- this provides a two-dimensional description of the main light- intensity changes in the visual output, including information about edges, contours, and BLOBS. · 2 1/2 D SKETCH- this incorporates a description of the depth and orientation of visible surfaces, making use of information provided by shadowing, texture, motion, binocular disparity and so on. Like the primal sketch, it is observer-centred or viewpoint-dependent. · 3-D MODEL REPRESENTATION- this described three-dimensionally the shaped of objects and their relative positions in a way that is independent of the observer’s viewpoint (VIEWPOINT-INVARIANT). Primal sketch According to MARR (1982), we can identify two versions of the primal sketch; the RAW PRIMAL SKETCH, and the FULL PRIMAL SKETCH. Both sketches are symbolic, meaning that they represent the image as a list of symbols. The RAW PRIMAL SKETCH contains information about light-intensity changes in the visual scene, and the FULL PRIMAL SKETCH makes use of this information to identify the number and outline shapes of visual objects. number and outline shapes of visual objects....
View Full Document
This note was uploaded on 05/04/2008 for the course PSYCH 2006 taught by Professor Newellf during the Winter '06 term at Trinity College Dublin.
- Winter '06