Automated digital document generation from digital videos
AssigneeAdobe Inc.
InventorsNiyati Himanshu Chhaya, Tripti Shukla, Jeevana Kruthi Karnuthala, Bhanu Prakash Reddy Guda, Ayudh Saxena, Abhinav Bohra, Abhilasha Sancheti, Aanisha Bhattacharyya
Life of this patent
Estimating the term…
Abstract
Techniques are described that support automated generation of a digital document from digital videos using machine learning. The digital document includes textual components that describe a sequence of entity and action descriptions from the digital video. These techniques are usable to generate a single digital document based on a plurality of digital videos as well as incorporate user-specified constraints in the generation of the digital document.
Claims20
A method of generating a digital document, the method comprising:
- receiving, by a computing device, at least one constraint to be used in controlling the generation of the digital document;
- locating, by the computing device, action clips that depict actions from at least one digital video;
- determining, by the computing device, frames from the action clips of the at least one digital video;
- forming, by the computing device, textual components that include entity and respective action descriptions based on the frames using a model trained using machine-learning;
- generating, by the computing device, a digital document based on the textual components and using the at least one constraint;
- and
- rendering the digital document by a display device.
A digital document generation system comprising:
- a processing system;
- a non-transitory computer readable media communicatively coupled to the processing system;
- a user interface supporting user inputs to specify at least one constraint that is used in controlling the generation of a digital document;
- an action detection module implemented by the processing system to locate action clips that depict actions from at least one digital video;
- a frame location module implemented by the processing system to locate frames from the at least one digital video;
- and
- a decoding module implemented by the processing system to form textual components to be used in generating the digital document, the textual components including a sequence of entity and respective action descriptions based on the located frames using a model trained using machine-learning.
A system comprising:
- means for inputting at least one constraint to be used in controlling generation of a digital document;
- means for locating action clips from at least one digital video;
- means for locating frames from the action clips of the at least one digital video;
- means for forming textual components that include a sequence of entity and respective action descriptions based on the located frames using a model trained using machine-learning;
- and
- means for generating the digital document based on the textual components, the means for generating the digital document using the at least one constraint in generation of the digital document.
Dependencies are inferred from the claim text.