Systems, devices and methods for distributed hierarchical video analysis
AssigneeHuawei Technologies Co., Ltd.
InventorsVarshanth Ravindra Rao, Peng Dai, Hanwen Liang, Md Ibrahim Khalil, Juwei Lu
Life of this patent
Estimating the term…
Abstract
System and method of analyzing a video, comprising dividing the video into a set of successive basic units; generating semantic tags for the basic units using a set of hierarchical classifier nodes that comprise a parent classifier node and a plurality of child classifier nodes, wherein the basic units are each routed through selected child classifier nodes based on classification of the basic units by the parent classifier node; and generating a semantic topic for the video based on the semantic tags generated for the basic units.
Claims18
A computer implemented method of analyzing a video, comprising:
- dividing the video into basic units, wherein the basic units are successive;
- generating semantic tags for the basic units using a set of classifier nodes, the set of classifier nodes comprising a set of hierarchical classifier nodes that have a parent classifier node and a plurality of child classifier nodes, wherein in generating the semantic tags for the basic units, the basic units are each routed through selected child classifier nodes based on one or more predefined routing logic;
- and
- generating a semantic topic for the video based on the semantic tags generated for the basic units.
A video analysis system comprising:
- one or more processor devices;
- and
- one or more memories storing machine-executable instructions, which when executed by the one or more processor devices, cause the system to:
- divide a video into basic units, wherein the basic units are successive;
- generate semantic tags for the basic units using a set of classifier nodes, the set of classifier nodes comprising a set of hierarchical classifier nodes that have a parent classifier node and a plurality of child classifier nodes, wherein in generating semantic tags for the basic units, the basic units are each routed through selected child classifier nodes based on one or more predefined routing logic;
- and
- generate a semantic topic for the video based on the semantic tags generated for the basic units.
A non-transitory computer readable medium storing software instructions for configuring an execution device to analyze a video by:
- dividing the video into basic units, wherein the basic units are successive;
- generating semantic tags for the basic units using a set of classifier nodes, the set of classifier nodes comprising a set of hierarchical classifier nodes that have a parent classifier node and a plurality of child classifier nodes, wherein in generating semantic tags for the basic units, the basic units are each routed through selected child classifier nodes based on one or more predefined routing logic;
- and
- generating a semantic topic for the video based on the semantic tags generated for the basic units.
Dependencies are inferred from the claim text.