Microsoft Technology Licensing, LLC

Jianlong Fu

Ji Li

Tao Mei

Huan Yang

Chin-Yew Lin

Jinpeng Wang

Yong Rui

Bei Liu

Kuiyuan Yang

Fatima Zohra Daha

Baining Guo

Houwen Peng

Various implementations of the subject matter relate to moment localization in media stream. In some implementations, a two-dimensional temporal feature map representing a plurality of moments within a media stream is extracted from the media stream, wherein the two-dimensional temporal feature map comprises a first dimension representing a start of a respective one of the plurality of moments and a second dimension representing an end of a respective one of the plurality of moments. A correlation between the plurality of moments and an action in the media stream is determined based on the two-dimensional temporal feature map.

Moment localization in media stream

There is provided a solution for image processing. In this solution, first and second information is determined based on texture features of an input image and a reference image. The first information at least indicates for a first pixel block in the input image a second pixel block in the reference image most relevant to the first pixel block in terms of the texture features, and the second information at least indicates a relevance of the first pixel block to the second pixel block. A transferred feature map with a target resolution is determined based on the first information and the reference image. The input image is transformed into an output image with the target resolution based on the transferred feature map and the second information. The output image reflects a texture feature of the reference image.

Image super-resolution

In embodiments of the present disclosure, there is provided a method for generating a stylized icon automatically. After a query text inputted by a user is obtained, a trained generator is used to generate a structured icon that can characterize a structure of an object, and then the structured icon is stylized, such as performing color padding or adding other styles, so as to generate a high-quality stylized icon for the user. In embodiments of the present disclosure, a structured icon and a stylized icon are generated respectively at two stages, where the structured icon can clearly characterize the structure of the object, while the stylized icon can be richer in color and style. Therefore, the stylized icon generated according to embodiments of the present disclosure has a higher quality and is more realistic, thereby improving the user experience of icon generation.

Automatic generation of stylized icons

A method and system for transforming an input image via a plurality of image transformation stylizers includes receiving the input image; providing the input image, information about the plurality of image transformation stylizers and at least one of user data, history data, and contextual data to a trained machine-learning (ML) model for selecting a subset of the plurality of image transformation stylizers; receiving as an output from the ML model the subset of image transformation stylizers; executing the subset of the image transformation stylizers on the input image to generate a plurality of transformed output images; ranking the plurality of transformed output images based on at least one of the input image, the user data, the history data, and the contextual data; and providing the ranked plurality of transformed output images for display.

Image transformation infrastructure

A system and method for rich content transformation are provided. The system and method allow rich content transformation to be separately processed on a client device and on a cloud-based server. The client device downsizes a rich content and transmits the downsized rich content to the cloud-based server via a network. The cloud-based server calculates function parameters based on the downsized rich content using one or more machine learning models included in the server. The calculated function parameters are transmitted to the client device via the network. The client device then applies these function parameters to the rich content on the client device to obtain the transformed rich content.

Efficient server-client machine learning solution for rich content transformation

In accordance with implementations of the subject matter described herein, a solution for fine-grained image recognition is proposed. This solution includes extracting a global feature of an image using a first sub-network of a first learning network; determining a first attention region of the image based on the global feature using a second sub-network of the first learning network, the first attention region including a discriminative portion of an object in the image; extracting a first local feature of the first attention region using a first sub-network of a second learning network; and determining a category of the object in the image based at least in part on the first local feature. Through this solution, it is possible to localize an image region at a finer scale accurately such that a local feature at a fine scale can be obtained for object recognition.

Fine-grained image recognition

Techniques and constructs to facilitate automatic tagging can provide improvements in image storage and searching. The constructs may enable training a deep network using tagged source images and target images. The constructs may also train a top layer of the deep network using a personal photo ontology. The constructs also may select one or more concepts from the ontology for tagging personal digital images.

Growing community of inventors

Beijing, China

Jianlong Fu