Abstract: A decoder includes: circuitry; and a memory connected to the circuitry, in which the circuitry, in operation, generates an image by decoding a bitstream, generates a plurality of intermediate feature maps having a uniform size based on the image, generates, based on the plurality of intermediate feature maps, a plurality of feature maps having different sizes for each hierarchical layer in a plurality of hierarchical layers included in a neural network for a machine task, and in generation of the feature maps, for at least one hierarchical layer of the plurality of hierarchical layers, selects any of: a first method of generating the plurality of feature maps of the at least one hierarchical layer using the plurality of intermediate feature maps, and a second method of generating the plurality of feature maps of the at least one hierarchical layer without using the plurality of intermediate feature maps but using the plurality of feature maps generated in a hierarchical layer different from the at least one h

Abstract: An image decoder includes circuitry and a memory connected to the circuitry. In operation, the circuitry decodes a first bitstream to acquire a first image, decodes a second bitstream to acquire specification information specifying a particular region in the first image and a second image containing image data for the particular region, and generates a third image based on the first image, the specification information, and the second image.

Abstract: A decoder includes circuitry and memory coupled to the circuitry. In operation, the circuitry: decodes a plurality of sets of neural network information each of which identifies a neural network filter; decodes, from one access unit, two or more sets of activation information each of which specifies one set of neural network information among the plurality of sets of neural network information; and applies, to one picture, two or more neural network filters identified by two or more sets of neural network information specified by the two or more sets of activation information.

Abstract: An image encoder includes circuitry, and a memory connected to the circuitry. The circuitry, in operation, determines a first region and a second region in an input image, generates a first image from the input image by applying a first prefilter to the first region and a second prefilter to the second region, determines a first parameter relating to the first region and a second parameter relating to the second region, and generates a bitstream by encoding the first region based on the first parameter, encoding the second region based on the second parameter, encoding the first prefilter or a first postfilter corresponding to the first prefilter, and encoding the second prefilter or a second postfilter corresponding to the second prefilter.

Abstract: An image decoding device includes circuitry and a memory connected to the circuitry, in which in operation, the circuitry acquires a first image and a plurality of first filter sets by decoding a bitstream, generates and outputs a second image by selecting one first filter set from the plurality of first filter sets based on usage information indicating an image usage, and applying the first filter set having been selected to the first image.

Abstract: An image encoding device includes: a filter processing unit that performs filter processing on an input image to generate a first image; and an encoding processing unit that generates a bit stream by performing encoding processing on the first image and transmits the bit stream to an image decoding device, in which the filter processing unit: includes a plurality of filters of different types; and selects one filter from the plurality of filters based on usage information indicating image usage on the image decoding device side and applies the selected filter to the input image.

Abstract: A decoder includes circuitry and memory coupled to the circuitry. In operation, the circuitry: decodes expression data indicating information expressed by a person; generates a person equivalent image corresponding to the person through a neural network according to the expression data and at least one profile image of the person; and outputs the person equivalent image.

Abstract: A battery module (102) is contained in a battery pack (100) and includes plural cells (118), a first cell holder (112), a second cell holder (114), and a current collector (116). The first cell holder (112) and the second cell holder (114) each have plural pockets to retain the cells (118). The first cell holder (112) and the second cell holder (114) are assembled to form a manifold to partially submerge the cells (118) within fluid. In addition, an interface (130) between a curved surface of the cells (118) and a curved surface of the pockets of at least one of the first cell holder (112) and the second cell holder (114) is sealed using a sealant. Further, the interface between the first cell holder (112) and the second cell holder (114) is sealed.

Abstract: A vehicle battery pack 100 includes at least one cell holder 102 for holding plural cells 106, and an enclosure 104 for enclosing the cell holder 102. The enclosure 106 includes a central body 112, a front endplate 114 and a rear endplate 116, each attached to a front and a rear of the central body 112, respectively. The central body 112 includes at least one coolant port 124 for filling a coolant in the enclosure 104. The central body 112 further includes a heat sink 126 for transferring heat from the coolant to the ambient area. The front endplate 114 includes a fan 130 and a pump 132, to make the battery pack 100 as a self-contained unit.

Abstract: An encoder includes circuitry and memory coupled to the circuitry. In operation, the circuitry encodes a video into a bitstream, the video is specified to be outputted at a first speed different from an original speed of the video, and in encoding the video, the circuitry encodes, into the bitstream, original speed information related to a second speed that is the original speed of the video.

Abstract: An encoder includes circuitry and memory coupled to the circuitry. In operation, the circuitry performs a first resampling process that changes a resolution of an image to be processed corresponding to one of an image to be encoded or a reference image for the image to be encoded, and in the first resampling process, the circuitry changes the resolution of the image to be processed, by changing a resolution of one component among a plurality of components of the image to be processed, and retains each resolution of one or more remaining components among the plurality of components.

Abstract: A method of streaming remotely located content, performed at a client device, comprises: communicating with a plurality of servers each of which hosts multiple copies of the content, said copies being encoded at different respective bitrates and each being divided into a plurality of segments that are arranged according to a time sequence; and requesting a concurrent download of a set of segments from a group of download servers that is at least a subset of the plurality of servers; wherein respective segments in the set are downloaded from different servers in the group of download servers, said segments being consecutive in the time sequence.

Abstract: A first path for forwarding an I/O request from a host device to a disk in a disk array is identified. The first path includes two endpoints (a first initiator endpoint on the host device and a first target endpoint on the disk array) separated by a storage area network. In response to an indication that the first path is non-functional, a second path to the disk for the I/O request is identified as an alternative to the first path. The second path includes a second initiator endpoint and a second target endpoint and is identified by selecting a path from among those paths that have at least one endpoint that is different from the two endpoints of the first path.

Abstract: A first path for forwarding an I/O request from a host device to a disk in a disk array is identified. The first path includes two endpoints (a first initiator endpoint on the host device and a first target endpoint on the disk array) separated by a storage area network. In response to an indication that the first path is non-functional, a second path to the disk for the I/O request is identified as an alternative to the first path. The second path includes a second initiator endpoint and a second target endpoint and is identified by selecting a path from among those paths that have at least one endpoint that is different from the two endpoints of the first path.