Register and login to view the CDVL license agreement. Send questions to the CDVL site administer, Margaret Pinson, at firstname.lastname@example.org
CDVL license agreement clarification (in plain English)
CDVL hosts video, audio, and images. We refer to all of these as “CDVL content.”
Research and Development (R&D) content
Most of the content hosted by CDVL is available for Research and Development (R&D) purposes only. The original copyright holder retains all other rights to the content.
The following uses are allowed for CDVL R&D content:
- To the extent that an authorized commercial entity is conducting internal research and development, use of the CDVL R&D content is appropriate.
- Sample frames from CDVL R&D content can be published in technical papers, scientific publications, standards contributions, and other standardization related documents.
- CDVL R&D content can be played at technology research and development events and standards meetings.
- Standards committees can use CDVL R&D content within subjective tests to validate objective video quality models (e.g., ATIS, VQEG, ITU, ISO, AOMedia, SMPTE, DVB, 3GPP).
- A commercial entity can use CDVL R&D content to develop a new product or to improve an existing product.
- A commercial entity can use CDVL R&D content for a free educational demonstration at a trade show, during a customer visit, or on their website. Care should be taken that no product marketing takes place during this free educational demonstration.
- CDVL R&D content can be modified (e.g., format converted, deinterlaced, converted to a different color space, trimmed, re-edited, dubbed with a different sound track, compressed, transcoded, transmitted, and corrupted with network impairments).
The following uses are not allowed for CDVL R&D content:
- Do not market a product
- Do not publish snapshots in product brochures
- Do not redistribute video with a commercial product
- Do not perform a subjective test for profit
- Do not use in television shows, commercials, or movies
- Do not distribute CDVL content (e.g., do not post to YouTube© or your website, do not give to a third party)
See the exception below for standards work, where a company can be sub-contracted to perform subjective tests or other evaluations for the SDO.
Public domain content
CDVL hosts a few sequences that are in the public domain. These public domain sequences are available for anyone to use for any purpose. These sequences are clearly marked as “public domain.”
Use of CDVL content by standards committees and commercial partnerships
The ITU, ISO, VQEG, ATIS, AOMedia, 3GPP, DVB, SMPTE, and other standards developing committees (SDO) can use CDVL R&D content for an internal project (e.g., a subjective test to validate objective video quality models).
- Each organization involved in the project must understand, agree to, and sign the CDVL database content user click-through agreement.
- This agreement appears when you try to access the "Find Videos" area. Sign digitally by agreeing to the terms and clicking "Continue".
- Afterward, the CDVL R&D content can be shared among project members for the purposes of that project.
This includes organizations that are peripherally involved (e.g., a university hired to do a subjective test for an SDO).
Other than the above exception, the usual CDVL usage terms apply. The CDVL R&D content (original and modified) does not need to be erased when the standards committee finishes that project. Each organization can archive the CDVL R&D content indefinitely and use it for other purposes.
Private R&D collaborations can use this same procedure (e.g., if a company and university collaborate to write a research paper, commercial company working under contract for a commercial partner). Please contact the CDVL Administrator, Margaret Pinson, for answers to questions not addressed here.
Redistribution of datasets with CDVL content
CDVL can host subjective datasets that contain video sequences derived from CDVL R&D content. Do not create a separate website to redistribute such datasets, unless permission is obtained by the content owners.
Frequently Asked Questions (FAQ)
Additional usage terms:
Some content have additional usage constraints, such as:
- Publications must also credit the content owner (e.g., in publications)
- An additional license applies (e.g., a particular open source license)
- Respect the privacy of people appearing in the content
These terms are listed in the description of the content on CDVL.
Please respect these constraints. CDVL content is provided on a no cost basis by the owners. The availability of such content depends upon CDVL users respecting the content providers’ requests.
Respect the privacy of people appearing in the content. You may recognize someone in a video sequence or image. Publications and discussions (public or private) must not identify that person by name.
Email addresses provided during CDVL registration will not be redistributed.
When CDVL content is used in a publication, please give credit to the CDVL website. This attribution helps us justify continued support and sponsorship of CDVL. The following paper reference is recommended:
Margaret H. Pinson, "The Consumer Digital Video Library [Best of the Web]," IEEE Signal Processing Magazine, vol. 30, no. 4, pp. 172,174, July 2013 doi: 10.1109/MSP.2013.2258265
For some content, publications must also give credit to the content owner. The description of the content on CDVL will mention this issue, if the content owner requires this acknowledgement.
Long video sequences
CDVL hosts some long sequences with audio.
- Search for "Fishin" to find "NTIA Fishin Florida" segments 1 to 6. These make a 3 minute sequence.
- Search for "Touch" to find "NTIA Touch Em' Up Boxing" segments 1 to 6. This makes a 3 minute sequence.
- Search for "Flamenco" to find "NTIA Flamenco Dancers" segments 1 to 9. This makes a 7 minute sequence.
- Search for "Atlanta" to find the 3D sequence "NTIA Atlanta Beltline" segments 1 to 12. If you use only one eye, this can be used as an HD sequence.
- Search for "Foot" to find "NTIA The Foot Music Video" segments 1 to 7. This makes a 4 minute sequence.
- Search for "SVT" to find the SVT sequence. This 6 minute sequence is available in a variety of formats.
- Search for "ElFluente" to find the 8 minute Netflix video.
- The "PSCR Emergency Telemedicine" dataset contains a variety of longer sequences that simulated emergency telemedicine response. This footage contains simulated injuries.
Public domain video sequences
CDVL hosts some public domain videos. The following searches will find these video sequences:
Match all keywords 'NTIA Purple Collage' in the title
Match all keywords 'NTIA Rainbow Collage' in the title
Match all keywords 'NTIA Burn Close-up' in the title
Match all keywords 'NTIA Collage Musikvid Settings' in the title
Match all keywords 'NTIA Colorado blue spruce (1e)' in the title
Match all keywords 'NTIA cropduster at sunrise' in the title
Match all keywords 'NTIA Duck Feed' in the title
Match all keywords 'NTIA Elephant Crane' in the title
Match all keywords 'NTIA Flatirons in Fall' in the title
Match all keywords 'NTIA Giraffe' in the title
Match all keywords 'NTIA Green Bird' in the title
Match all keywords 'NTIA intersection on a snowy day (6a)' in the title
Match all keywords 'NTIA Lion' in the title
Match all keywords 'NTIA Red Parrot' in the title
Match all keywords 'NTIA Sprinkler' in the title
Match all keywords 'NTIA The Cloud' in the title
Match all keywords 'NTIA Timelapse of Red Prayer Plant and Peacock Plant' in thetitle
Match all keywords 'NTIA Wind Public Domain' in the title
Match all keywords 'NTIA Zoocuts' in the title
- Select dataset name 'ANSI T1.801.01'
- Select dataset name 'ITU-R Rec. BT.802, 525-line'
- Select dataset name 'ITU-R REc. BT.802, 625-line'
CDVL went live on Wednesday, November 4, 2009. CDVL's creation was a three year joint effort by:
- Intel Corporation (Technical Research)
- The University of California at Santa Barbara (Department of Electrical and Computer Engineering)
NTIA/ITS maintains CDVL and contributed thousands of video scenes. NTIA/ITS regularly makes more content available.