Microsoft Technology Licensing, LLC

Yinpeng Chen

Zicheng Liu

Zhengyou Zhang

Dongdong Chen

Lu Yuan

Mengchen Liu

Xiyang Dai

Sasa Junuzovic

Nikolaos Karianakis

Kun Luo

Ying Jin

Philip Andrew Chou

Alastair Wolman

Cha Zhang

Dinei A Florencio

Lei Zhang

Hongli Deng

Mei Chen

Sean Eron Anderson

Ye Yu

Pei Qun Yu

Lijuan Wang

Bin Xiao

Yunsheng Li

Pei Yu

A method for environment-specific training of a machine learning model, comprises receiving, for a local environment, a data stream including a plurality of sequential data snippets. Programmed labels are generated for each data snippet using a student version of a machine learning model. A portion of data snippets and associated programmed labels are selected and uploaded to a server-side computing device for evaluation by a teacher version of the machine learning model. An environment-specific training update is received from the server-side computing device. This training update is based on a comparison of the selected programmed labels and pseudolabels generated for the selected portion of data snippets by the teacher version. The environment-specific training update is applied to the student version to generate an updated student version. The updated student version of the machine learning model is then used to generate programmed labels for newly received data snippets.

Environment-specific training of machine learning models

A computer device for automatic feature detection comprises a processor, a communication device, and a memory configured to hold instructions executable by the processor to instantiate a dynamic convolution neural network, receive input data via the communication network, and execute the dynamic convolution neural network to automatically detect features in the input data. The dynamic convolution neural network compresses the input data from an input space having a dimensionality equal to a predetermined number of channels into an intermediate space having a dimensionality less than the number of channels. The dynamic convolution neural network dynamically fuses the channels into an intermediate representation within the intermediate space and expands the intermediate representation from the intermediate space to an expanded representation in an output space having a higher dimensionality than the dimensionality of the intermediate space. The features in the input data are automatically detected based on the expanded representation.

Dynamic matrix convolution with channel fusion

The disclosure herein describes generating an inpainted image from a masked image using a patch-based encoder and an unquantized transformer. An image including a masked region and an unmasked region is received, and the received image is divided into a plurality of patches including masked patches. The plurality of patches is encoded into a plurality of feature vectors, wherein each patch is encoded to a feature vector. Using a transformer, a predicted token is generated for each masked patch using a feature vector encoded from the masked patch, and a quantized vector of the masked patch is determined using generated predicted token and a masked patch-specific codebook. The determined quantized vector of the masked patch is included into a set of quantized vectors associated with the plurality of patches, and an output image is generated from the set of quantized vectors using a decoder.

Generating an inpainted image from a masked image using a patch-based encoder

To improve the accuracy and efficiency of object detection through computer digital image analysis, the detection of some objects can inform the sub-portion of the digital image to which subsequent computer digital image analysis is directed to detect other objects. In such a manner object detection can be made more efficient by limiting the image area of a digital image that is analyzed. Such efficiencies can represent both computational efficiencies and communicational efficiencies arising due to the smaller quantity of digital image data that is analyzed. Additionally, the detection of some objects can render the detection of other objects more accurate by adjusting confidence thresholds based on the detection of those related objects. Relationships between objects can be utilized to inform both the image area on which subsequent object detection is performed and the confidence level of such subsequent object detection.

Focused computer detection of objects in images

Systems and methods for object detection generate a feature pyramid corresponding to image data, and rescaling the feature pyramid to a scale corresponding to a median level of the feature pyramid, wherein the rescaled feature pyramid is a four-dimensional (4D) tensor. The 4D tensor is reshaped into a three-dimensional (3D) tensor having individual perspectives including scale features, spatial features, and task features corresponding to different dimensions of the 3D tensor. The 3D tensor is used with a plurality of attention layers to update a plurality of feature maps associated with the image data. Object detection is performed on the image data using the updated plurality of feature maps.

Dynamic head for object detection

An object re-identifier. For each of a plurality of frames of a video, a quality of the frame is assessed and a confidence that a previously-recognized object is present in the frame is determined. The determined confidence for the frame is weighted based on the assessed quality of the frame such that frames with higher relative quality are weighted more heavily than frames with lower relative quality. An overall confidence that the previously-recognized object is present in the video is assessed based on the weighted determined confidences.

Depth-based object re-identification

The large display interaction implementations described herein combine mobile devices with people tracking to enable new interactions including making a non-touch-sensitive display touch-sensitive and allowing personalized interactions with the display. One implementation tracks one or more mobile computing device users relative to a large computer-driven display, and configures content displayed on the display based on a distance a given mobile computing device user is from the display. Another implementation personalizes user interactions with a large display. One or more mobile computing device users are tracked relative to a display. The identity of each of the one or more mobile computing device users is obtained. Content displayed on the display is configured based on a distance an identified mobile computing device user is from the display and the identity of the user that provides the content.

Combining mobile devices with people tracking for large display interactions

Systems and methods for person counting are disclosed. A method may include retrieving an image frame from a plurality of image frames captured by a camera. The image frame may be split into a grid of a plurality of cells of a pre-determined cell dimensions. The pre-determined cell dimensions may be based on dimensions of the retrieved image frame and reference dimensions of training images of a person detection classifier. At least a portion of the plurality of cells may be rearranged to generate a new image. The new image may be padded with at least one padding strip to adjust dimensions of the new image to the reference dimensions of the training images. Person detection may be performed using the new image and the person detection classifier to obtain a number of persons detected within the new image.

System and method for person counting in image data

The description relates to interactions with a display device. In one example, the interactions can include detecting a user proximate to a display and detecting a non-touch control gesture performed by the user proximate to the display. The example can also include presenting a graphical user interface (GUI) on the display that includes options associated with the control gesture. The example can also include receiving user input selecting one of the options and receiving additional user input from the user to interact with the GUI via the selected one of the options.

Device control

The description relates to remote collaboration via a telepresence experience. One example can include an interactive digital display. The example can also include a virtual user presentation component configured to generate a graphical user interface that includes a virtual representation of a remote user on the interactive digital display. The graphical user interface can be configured to present the remote user in a side by side or mirror image relationship to a local user of the interactive digital display.

Telepresence experience

Techniques for automatically connecting to a service controller are described herein. In one example, a service controller device includes a processor and a computer-readable memory storage device storing executable instructions that cause the processor to broadcast at least one of an access credential, connection information or an access credential hash embedded in an audio signal. The processor can also authenticate a client device based on a transmission of at least one of the connection information, the access credential, or the access credential hash from the client device to the client connector and transmit data to the client device in response to authenticating the client device.

Growing community of inventors

Sammamish, WA, United States of America

Yinpeng Chen