1 / 4

Proposed Guidelines for Research Publications using SVS Customer Data

Proposed Guidelines for Research Publications using SVS Customer Data. Rick Kjeldsen Sharath Pankanti. Problem. Research team is asked to make extensive use of SVS customer data Real customer data presents different problems than “similar” data in the public domain

korene
Télécharger la présentation

Proposed Guidelines for Research Publications using SVS Customer Data

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Proposed Guidelines for Research Publications using SVS Customer Data Rick Kjeldsen SharathPankanti

  2. Problem • Research team is asked to make extensive use of SVS customer data • Real customer data presents different problems than “similar” data in the public domain • Often, the most important research addresses these differences • Researchers are evaluated in part on their academic publication record • Publication requires example images • Publication requires performance results based on image data • Publication timelines do not permit an extensive review process BUT • Customer data must be kept confidential, specifically information about where security cameras are located and how to defeat them

  3. Proposed guidelines • Image data will include no information which could reasonably be used to identify the location. This will include • Street and other signs or markings that could identify the location • Recognizable structures or road configurations • Long views where city-scapes may be recognizable • Performance data will be anonymous with respect to data source • Problems will be discussed in generic context and will not reveal any proprietary (e.g., use case, market sector) information. • Problems specific to one customer will not be discussed, even anonymously • Specific techniques for defeating the system or avoiding detection will not be discussed. • Images where individuals are recognizable will not be used without their permission • There will be no reference to specific customers or code release versions • SVS / VCAS will not be mentioned by name • Proprietary technology must be protected before publication

  4. Examples • The following are examples of image data which are sufficiently anonymous that they may appear in a publication • The following web images illustrate cameras that are not sufficiently anonymous to be used

More Related