内容

目次

Object recognition: shape-based recognition; what is recognition? whyobject recognition is difficult. Approaches to object recognition: invariantproperties and feature spaces; parts and structural descriptions; thealignment approach; which is the correct approach?. The alignment ofpictorial descriptions: using corresponding features; the use of multiplemodels for 3-D objects; aligning pictorial descriptions; transforming theimage or the models? before and after alignment. The alignment of smoothbounding contours: the curvate method; accuracy of the curvature method;empirical testing. Recognition by the combination of views: modelling objectsby view combinations; objects with sharp edges; using two views only; using asingle view; the use of depth values; summary of the basic scheme; objectswith smooth boundaries; recognition by image combinations; extensions to theview-combination scheme; psychophysical and physiological evidence; interimconclusions: recognition by multiple views. Classifications: classificationand identification; the role of object classification; class-basedprocessing; using class prototypes; pictorial classification; evidence frompsychology and biology; are classes in the world or in our head? theorganization of recognition memory. Image and model correspondence: featurecorrespondence; contour matching; correspondence-less methods; correspondenceprocesses in human vision; model construction; compensating for illuminationchanges. Segmentation and saliency: is segmentation feasible? bottom-up andtop-down segmentation; extracting globally salient structures; saliency,selection, and completion; what can bottom-up segmentation achieve? Visualcognition and visual routines: perceiving "inside" and "outside"; spatialanalysis by visual routines; conclusions and open problems; the elementaloperations; the assembly and storage of routines; routines and recognition.Sequence seeking and counter streams - a model for visual cortex: thesequence-seeking scheme; biological embodiment; summary. Appendices:alignment by feature; the curvature method; errors of the curvature method;locally affine matching; definitions.