Abstract: An object, such as a soccer ball, is provided with perforated, interior panels disposed between an interior inflatable bladder and an outer covering which is stitched together from a plurality of individual panels. The stitching process produces raised portions surrounding the interior sides of the outer covering. These raised portions define recesses into which the perforated interior panels are disposed. The recess-filling panels overlie the ridges which are hard and whose hardness is counteracted by these preferably perforated panels. This provides a smooth surface against which the inflatable bladder rests. Additional flexure is also provided by perforating one or more of the layers in the outer covering.

Abstract: An object, such as a soccer ball or a handball, is provided with perforated, interior panels disposed between an interior inflatable bladder and an outer covering which is stitched together from a plurality of individual panels. The stitching process produces raised portions surrounding the interior sides of the outer covering. These raised portions define recesses into which the perforated interior panels are disposed. The recess-filling panels overlie the ridges which are hard and whose hardness is counteracted by these preferably perforated panels. This provides a smooth surface against which the inflatable bladder rests. Additional flexure is also provided by perforating one or more of the layers in the outer covering. With particular reference to handball manufacture the interior panels comprise two layers, the outer one of which contains recesses containing shock absorbing gel type material.

Abstract: A method re-identifies objects in a pair of images by applying a convolutional neural network (CNN). Each layer in the network operates on an output of a previous layer. The layers include a first convolutional layer and a first max pooling layer to determine a feature map, a cross-input neighborhood differences layer to produce neighborhood difference maps, a patch summary layer to produce patch summary feature maps, a first fully connected layer to produce a feature vector representing higher order relationships in the patch summary feature maps, a second fully connected layer to produce two scores representing positive pair and negative pair classes, and a softmax layer to produce positive pair and negative pair probabilities. Then, the positive pair probability is output to signal whether the two images represent the same object or not.

Abstract: Techniques are disclosed herein that enable digital images to be segmented based on a user's semantic input. In other words, given an input image of a person walking a dog adjacent to a tree, a user can simply provide the semantic input “dog” and the system will segment the dog from the other elements in the image. If the user provides other semantic input, such as “person” or “tree”, the system will segment the person or the tree, respectively, from the same image. Using semantic input advantageously eliminates any need for a user to directly interact with the input image through a tedious process of painting brush strokes, tracing boundaries, clicking target points, and/or drawing bounding boxes. Thus semantic input represents an easier and more intuitive way for users to interact with an image segmentation interface, thereby enabling novice users to take advantage of advanced image segmentation techniques.

Abstract: Techniques are disclosed herein that enable digital images to be segmented based on a user's semantic input. In other words, given an input image of a person walking a dog adjacent to a tree, a user can simply provide the semantic input “dog” and the system will segment the dog from the other elements in the image. If the user provides other semantic input, such as “person” or “tree”, the system will segment the person or the tree, respectively, from the same image. Using semantic input advantageously eliminates any need for a user to directly interact with the input image through a tedious process of painting brush strokes, tracing boundaries, clicking target points, and/or drawing bounding boxes. Thus semantic input represents an easier and more intuitive way for users to interact with an image segmentation interface, thereby enabling novice users to take advantage of advanced image segmentation techniques.

Abstract: Techniques are disclosed herein that enable digital images to be segmented based on a user's semantic input. In other words, given an input image of a person walking a dog adjacent to a tree, a user can simply provide the semantic input “dog” and the system will segment the dog from the other elements in the image. If the user provides other semantic input, such as “person” or “tree”, the system will segment the person or the tree, respectively, from the same image. Using semantic input advantageously eliminates any need for a user to directly interact with the input image through a tedious process of painting brush strokes, tracing boundaries, clicking target points, and/or drawing bounding boxes. Thus semantic input represents an easier and more intuitive way for users to interact with an image segmentation interface, thereby enabling novice users to take advantage of advanced image segmentation techniques.

Abstract: Techniques are disclosed herein that enable digital images to be segmented based on a user's semantic input. In other words, given an input image of a person walking a dog adjacent to a tree, a user can simply provide the semantic input “dog” and the system will segment the dog from the other elements in the image. If the user provides other semantic input, such as “person” or “tree”, the system will segment the person or the tree, respectively, from the same image. Using semantic input advantageously eliminates any need for a user to directly interact with the input image through a tedious process of painting brush strokes, tracing boundaries, clicking target points, and/or drawing bounding boxes. Thus semantic input represents an easier and more intuitive way for users to interact with an image segmentation interface, thereby enabling novice users to take advantage of advanced image segmentation techniques.

Abstract: The invention pertains to a composition of a ketone peroxide comprising a) a peroxide derivative of the formula HOO—C(R1)(R2)—OOH wherein R1 is a branched or unbranched alkyl group with 1 to 4 carbon atoms or alkenyl group with 2 to 4 carbon atoms; and R2 is a branched or unbranched alkyl or alkenyl group with 5 to 12 carbon atoms; and b) a branched or unbranched hydrocarbon solvent; the peroxide derivative of a) having a solubility more than 40 g in 100 g of the solvent of b) at 20° C.; and comprises less than 10 wt % of a peroxide derivative of the formula HOO—C(R1)(R2)—OO—C(R1)(R2)—OOH, wherein R1 and R2have the previously given meanings.

Abstract: Methyl propyl ketone peroxide formulations are claimed that can be safely handled at ambient temperature, are storage stable, and give fast curing rates of unsaturated polyester resins particularly when accelerated with a complex of two different metals. The claimed products are particularly suited for thin articles and coating such as suited for use in gel coat systems. The pure ketone peroxides consist of more than 60% by weight of type-4 methyl propyl ketone peroxide and used as formulations containing sufficient amount of a non-aqueous phlegmatizer.

Abstract: Methyl propyl ketone peroxide formulations are claimed that can be safely handled at ambient temperature, are storage stable, and give fast curing rates of unsaturated polyester resins particularly when accelerated with a complex of two different metals. The claimed products are particularly suited for thin articles and coating such as suited for use in gel coat systems. The pure ketone peroxides consist of more than 60% by weight of type-4 methyl propyl ketone peroxide and used as formulations containing sufficient amount of a non-aqueous phlegmatizer.

Abstract: The present invention relates to a new class of peroxides and to a process for the preparation of these peroxides having the general formula (I), wherein n=1 or 2, R1, R2, R4, R5, and R6 are independently selected from the group comprising hydrogen, C1-C20 alkyl, C3-C20 cycloalkyl, C6-C20 aryl, C7-C20 aralkyl, and C7-C20 alkaryl, or R1 and R2 form a C3-C12 cycloalkyl group, which groups may include linear or branched alkyl moieties; and each of R1, R2, R4, R5, and R6 may optionally be substituted with one or more groups selected from hydroxy, alkoxy, linear or branched alkyl, aryloxy, halogen, ester, carboxy, nitrile, and amido, and R1 and R2 may form a ring, and R3 is independently selected from the group comprising C1-C20, alkyl; C3-C20 cycloalkyl, C6-C20 aryl, C7-C20, aralkyl, and C7-C20 alkaryl, which groups may include linear or branched alkyl moieties; and R3 may optionally be substituted with one or more groups selected from hydroxy, alkoxy, linear or branched alkyl, aryloxy,

Abstract: A plurality of targets comprise two spaced metal screens secured in electrical isolation. A voltage is applied to the screens such that when the screens are shorted by a penetrating metal projectile, a pulse is produced having a pulse duration representing a particular type of projectile according to the length of the projectile. A logic circuit including a counter produces a count signal representing the pulse duration and applies this count signal to a CPU which compares the count signal to stored time intervals to produce a projectile identifying signal which is applied to a LCD display for describing the type of projectile hitting the target. A target signal generator generates a signal representing any one of the targets hit in response to a hit on that one target and supplies that signal to the CPU which generates a display signal identifying the target descriptively.