Techniques for generating magnetic resonance (MR) images of a subject from MR data obtained by a magnetic resonance imaging (MRI) system, the techniques including: obtaining input MR data obtained by imaging the subject using the MRI system; generating a plurality of transformed input MR data instances by applying a respective first plurality of transformations to the input MR data; generating a plurality of MR images from the plurality of transformed input MR data instances and the input MR data using a non-linear MR image reconstruction technique; generating an ensembled MR image from the plurality of MR images at least in part by: applying a second plurality of transformations to the plurality of MR images to obtain a plurality of transformed MR images; and combining the plurality of transformed MR images to obtain the ensembled MR image; and outputting the ensembled MR image.

Methods are provided for the automated detection and/or counting of micro-objects in a microfluidic device. In addition, methods are provided for repositioning micro-objects in a microfluidic device. In addition, methods are provided for separating micro-objects in a spatial region of the microfluidic device.

The present disclosure relates to methods and apparatus for image processing. The apparatus can generate object mask information for one or more objects in a first image of a plurality of images in a scene. In some aspects, the first image can be at least one of a downscaled image, a down-sampled image, or a low resolution image. The apparatus can also determine one or more object classifications of the first image based on the generated object mask information. Additionally, the apparatus can identify a modification to at least one of the one or more object classifications based on a second image of the plurality of images in the scene. In some aspects, the apparatus can adjust or maintain the one or more object classifications based on the identified modification to at least one of the one or more object classifications.

An apparatus, method and computer program wherein the apparatus comprises:a plurality of quantum dot-graphene field effect transistors; circuitry configured to provide an individual drain-source bias voltage to each of a plurality of quantum dot-graphene field effect transistors, wherein different individual drain-source bias voltages have different parameters, to enable the plurality of quantum dot-graphene field effect transistors to detect light from a user of an apparatus; and circuitry configured to obtain output signals from each of a plurality of quantum dot-graphene field effect transistors where the output signal is dependent upon both the light detected by the quantum dot-graphene field effect transistor and the parameters of the drain-source bias voltage to enable the obtained output signals to be used as a scrambled identification signal of the user of the apparatus.

The disclosure provides a pedestrian re-identification method based on a spatio-temporal joint model of a residual attention mechanism and a device thereof. The method includes: performing feature extraction for an input pedestrian with a pre-trained ResNet-50 model; constructing a residual attention mechanism network including a residual attention mechanism module, a feature sampling layer, a global average pooling layer and a local feature connection layer; calculating a feature distance by using a cosine distance and denoting the feature distance as a visual probability according to the trained residual attention mechanism network; performing modeling for a spatio-temporal probability according to camera ID and frame number information in a pedestrian tag of a training sample, and performing Laplace smoothing for a probability model; and calculating a final spatio-temporal joint probability by using the visual probability and the spatio-temporal probability to obtain a pedestrian re-identification result.

A mask structure optimization device includes a classification target image size acquisition unit that is configured to acquire a size of a classification target image which is an image including a classification target, a mask size setting unit that is configured to set a size of a mask applied to the classification target image, a brightness detection unit that is configured to detect a brightness of each pixel within the classification target image at a position on an opposite side of the mask from the classification target image, a sum total brightness calculation unit that is configured to calculate the sum total brightness of the each pixel within the classification target image detected by the brightness detection unit, an initial value setting unit that is configured to set an initial value for a mask pattern of the mask, and a movement unit that is configured to relatively move the mask with respect to the classification target image. The sum total brightness calculation unit is configured to calculate the sum total brightness of the each pixel within the classification target image every time the movement unit relatively moves the mask by a predetermined movement amount. The mask structure optimization device further includes a mask pattern optimization unit that is configured to optimize the mask pattern of the mask on the basis of the sum total brightness.

The invention provides the techniques and systems that allow for the identification and classification of objects (i.e., humans) in images using a predictive segmentation model. More specifically, human forms are identified within an image by generating pixel-level bounding boxes for each possible object and using offsets and segmentation masking. In some instances, embodiments of the invention use an identified floor plane that intersects with a bounding box to identify a three-dimensional position for the intersection point, which can then be assigned to the human form and represent its depth within the image.

A device comprising a circuitry configured to obtain a sequence of digital images from an image sensor; select a region of interest within a digital image of the sequence of digital images; perform motion compensation on the region of interest to obtain a motion compensated region of interest based on motion information obtained from the sequence of digital images and a predefined accumulated time interval; define a mask pattern based on the compensated region of interest; apply the mask pattern to an electronic light valve.

A system obtains an image from a digital image stream captured by an imaging component. Both a foreground region of interest and a background region of interest present in the obtained image are identified, and the imaging component is zoomed out as appropriate to maintain a margin (a number of pixels) around both the foreground region of interest and the background region of interest. Additionally, a position of regions of interest (e.g., the background region of interest and the foreground region of interest) to improve the composition or aesthetics of the image is determined, and a composition score of the image indicating how good the determined position is from an aesthetics point of view is determined. A zoom adjustment value is determined based on the position of the regions of interest, and the imaging component is caused to zoom in or out in accordance with the zoom adjustment value.

A machine learning system can generate an image mask (e.g., a pixel mask) comprising pixel assignments for pixels. The pixels can he assigned to classes, including, for example, face, clothes, body skin, or hair. The machine learning system can be implemented. using a convolutional neural network that is configured to execute efficiently on computing devices having limited resources, such as mobile phones. The pixel mask can be used to more accurately display video effects interacting with a user or subject depicted in the image.