Adobe, Inc.

Varun Dua

Deepak Kumar

Ankur Jain

Suvrat Chaturvedi

Vikas Patidar

Saswata Dutta

The conversion quality evaluation system may evaluate the conversion quality of digital forms digitized from paper forms. The system may attribute multiple labels and respective confidence scores to form objects in a form, and aggregate information of form objects to form a conversion index, which is a measurement of conversion quality of the form. The conversion index may comprise various characteristics of the form, including cost of correction, accessibility, content reusability, or device readiness. As such, the conversion quality evaluation system can sort digital forms based on their conversion quality and provide guidance for post-digitization activities.

Conversion quality evaluation for digitized forms

Electronic form creation techniques are disclosed which automatically recommend candidate form objects to include in an electronic form being created. In some examples, a method may include receiving a request to create an electronic form, identifying a preceding N form objects created in the electronic form, identifying a candidate form object based on the identified preceding N form objects and one or more relationships between pre-existing form objects, and recommending the candidate form object for creation in the electronic form. The pre-existing form objects are included in multiple pre-existing forms. The method may further include identifying the one or more relationships between pre-existing form objects. The pre-existing forms may be selected, for example, based on information associated with the request (e.g., form type, an ID indicating identity of author creating the form, and/or ID indicating identity of a group to which the form author belongs).

Intelligent form creation

Electronic form generation techniques are disclosed which automatically generate an electronic form based on a paper form by semantically mapping form fields. An example method may include identifying a data field in a paper form, extracting a semantic meaning of the data field, searching multiple data elements to identify a data element that semantically matches the data field, and, responsive to identifying a semantically matching data element, generating a data field in the electronic form based on the semantically matched data element and the associated one or more digital data capture constructs. Each data element may have associated therewith one or more digital data capture constructs. The generated data field in the electronic form corresponds to the data field in the paper form.

Growing community of inventors

Bengaluru, India

Varun Dua