公开(公告)号：US12223412B2

公开(公告)日：2025-02-11

申请号：US17123697

申请日：2020-12-16

Applicant: Microsoft Technology Licensing, LLC

Inventor： Yinpeng Chen ,  Xiyang Dai ,  Mengchen Liu ,  Dongdong Chen ,  Lu Yuan ,  Zicheng Liu ,  Ye Yu ,  Mei Chen ,  Yunsheng Li

IPC: G06N3/04 ,  G06F17/16 ,  G06V10/70

Abstract: 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.

公开(公告)号：US12192543B2

公开(公告)日：2025-01-07

申请号：US18393664

申请日：2023-12-21

Applicant: Microsoft Technology Licensing, LLC

Inventor： Gaurav Mittal ,  Ye Yu ,  Mei Chen ,  Junwen Chen

IPC: H04N21/23 ,  G06T7/246 ,  G06V20/40 ,  H04N21/234

Abstract: Example solutions for video frame action detection use a gated history and include: receiving a video stream comprising a plurality of video frames; grouping the plurality of video frames into a set of present video frames and a set of historical video frames, the set of present video frames comprising a current video frame; determining a set of attention weights for the set of historical video frames, the set of attention weights indicating how informative a video frame is for predicting action in the current video frame; weighting the set of historical video frames with the set of attention weights to produce a set of weighted historical video frames; and based on at least the set of weighted historical video frames and the set of present video frames, generating an action prediction for the current video frame.

公开(公告)号：US11544561B2

公开(公告)日：2023-01-03

申请号：US16875782

申请日：2020-05-15

Applicant: Microsoft Technology Licensing, LLC

Inventor： Gaurav Mittal ,  Victor Manuel Fragoso Rojas ,  Nikolaos Karianakis ,  Mei Chen ,  Chang Liu

IPC: G06N3/08 ,  G06K9/62

Abstract: Providing a task-aware recommendation of hyperparameter configurations for a neural network architecture. First, a joint space of tasks and hyperparameter configurations are constructed using a plurality of tasks (each of which corresponds to a dataset) and a plurality of hyperparameter configurations. The joint space is used as training data to train and optimize a performance prediction network, such that for a given unseen task corresponding to one of the plurality of tasks and a given hyperparameter configuration corresponding to one of the plurality of hyperparameter configurations, the performance prediction network is configured to predict performance that is to be achieved for the unseen task using the hyperparameter configuration.

公开(公告)号：US12299082B2

公开(公告)日：2025-05-13

申请号：US18599029

申请日：2024-03-07

Applicant: MICROSOFT TECHNOLOGY LICENSING, LLC

Inventor： Gaurav Mittal ,  Nikolaos Karianakis ,  Victor Manuel Fragoso Rojas ,  Mei Chen ,  Jedrzej Jakub Kozerawski

IPC: G06F18/2431 ,  G06N3/04 ,  G06N3/08

Abstract: A method of balancing a dataset for a machine learning model includes identifying confusing classes of few-shot classes for a machine learning model during validation. One of the confusing classes and an image from one of the few-shot classes are selected. An image perturbation is computed such that the selected image is classified as the selected confusing class. The selected image is modified with the computed perturbation. The modified selected image is added to a batch for training the machine learning model.

公开(公告)号：US11443455B2

公开(公告)日：2022-09-13

申请号：US16744068

申请日：2020-01-15

Applicant: Microsoft Technology Licensing, LLC

Inventor： Victor M. Fragoso Rojas ,  Mei Chen ,  Gabriel Takacs

IPC: G06T7/73 ,  H04N5/247 ,  G06T7/60 ,  G06T7/80

Abstract: A scale and pose estimation method for a camera system is disclosed. Camera data for a scene acquired by the camera system is received. A rotation prior parameter characterizing a gravity direction is received. A scale prior parameter characterizing scale of the camera system is received. A cost of a cost function is calculated for a similarity transformation that is configured to encode a scale and pose of the camera system. The cost of the cost function is influenced by the rotation prior parameter and the scale prior parameter. A solved similarity transformation is determined upon calculating a cost for the cost function that is less than a threshold cost. An estimated scale and pose of the camera system is output based on the solved similarity transformation.

公开(公告)号：US11895343B2

公开(公告)日：2024-02-06

申请号：US17852310

申请日：2022-06-28

Applicant: Microsoft Technology Licensing, LLC

Inventor： Gaurav Mittal ,  Ye Yu ,  Mei Chen ,  Junwen Chen

IPC: H04N21/23 ,  H04N21/234 ,  G06V20/40 ,  G06T7/246

CPC classification number: H04N21/23418 ,  G06T7/246 ,  G06V20/46 ,  G06T2207/10021

Abstract: Example solutions for video frame action detection use a gated history and include: receiving a video stream comprising a plurality of video frames; grouping the plurality of video frames into a set of present video frames and a set of historical video frames, the set of present video frames comprising a current video frame; determining a set of attention weights for the set of historical video frames, the set of attention weights indicating how informative a video frame is for predicting action in the current video frame; weighting the set of historical video frames with the set of attention weights to produce a set of weighted historical video frames; and based on at least the set of weighted historical video frames and the set of present video frames, generating an action prediction for the current video frame.

公开(公告)号：US12087043B2

公开(公告)日：2024-09-10

申请号：US17535517

申请日：2021-11-24

Applicant: Microsoft Technology Licensing, LLC

Inventor： Gaurav Mittal ,  Ye Yu ,  Mei Chen ,  Jay Sanjay Patravali

IPC: G06K9/00 ,  G06F16/73 ,  G06F16/75 ,  G06N20/00 ,  G06V10/764 ,  G06V10/774

CPC classification number: G06V10/7753 ,  G06F16/73 ,  G06F16/75 ,  G06N20/00 ,  G06V10/764 ,  G06V10/7747

Abstract: The disclosure herein describes preparing and using a cross-attention model for action recognition using pre-trained encoders and novel class fine-tuning. Training video data is transformed into augmented training video segments, which are used to train an appearance encoder and an action encoder. The appearance encoder is trained to encode video segments based on spatial semantics and the action encoder is trained to encode video segments based on spatio-temporal semantics. A set of hard-mined training episodes are generated using the trained encoders. The cross-attention module is then trained for action-appearance aligned classification using the hard-mined training episodes. Then, support video segments are obtained, wherein each support video segment is associated with video classes. The cross-attention module is fine-tuned using the obtained support video segments and the associated video classes. A query video segment is obtained and classified as a video class using the fine-tuned cross-attention module.

公开(公告)号：US12056896B2

公开(公告)日：2024-08-06

申请号：US17931238

申请日：2022-09-12

Applicant: Microsoft Technology Licensing, LLC

Inventor： Victor M. Fragoso Rojas ,  Mei Chen ,  Gabriel Takacs

IPC: G06T7/73 ,  G06T7/60 ,  G06T7/70 ,  G06T7/80 ,  G06T17/00 ,  H04N23/90

CPC classification number: G06T7/75 ,  G06T7/60 ,  G06T7/70 ,  G06T7/80 ,  G06T17/00 ,  H04N23/90 ,  G06T2207/30244

Abstract: A scale and pose estimation method for a camera system is disclosed. Camera data for a scene acquired by the camera system is received. A scale prior parameter characterizing scale of the camera system is received. A cost of a cost function is calculated for a similarity transformation. The cost of the cost function is influenced at least by the scale prior parameter. Based at least on the cost function being less than a threshold cost, an estimated scale and pose of the camera system is output based on the similarity transformation.

公开(公告)号：US11960574B2

公开(公告)日：2024-04-16

申请号：US17361146

申请日：2021-06-28

Applicant: MICROSOFT TECHNOLOGY LICENSING, LLC

Inventor： Gaurav Mittal ,  Nikolaos Karianakis ,  Victor Manuel Fragoso Rojas ,  Mei Chen ,  Jedrzej Jakub Kozerawski

IPC: G06F18/2431 ,  G06N3/04 ,  G06N3/08

CPC classification number: G06F18/2431 ,  G06N3/04 ,  G06N3/08

Abstract: A method of balancing a dataset for a machine learning model includes identifying confusing classes of few-shot classes for a machine learning model during validation. One of the confusing classes and an image from one of the few-shot classes are selected. An image perturbation is computed such that the selected image is classified as the selected confusing class. The selected image is modified with the computed perturbation. The modified selected image is added to a batch for training the machine learning model.