microsoft/p.910

Crowdsourced implementation of P.910

P.910 Crowd

The P.910-Crowd Toolkit is a software package that enables users to run subjective video quality assessment test in a crowdsourcing platform like Amazon Mechanical Turk (AMT) or remote testing with a dedicated panel of workers. It is intended to be counter part of the ITU-T Recommendation P.910 for crowdsourcing.

Note: Currently full-screen play-back is not supported by HITs builds inside AMT. Therefore, we provide, the HITAPP Server which can serve the experiment. Consequently you can use AMT or any other crowdsourcing platform and post your study URL.

For more information about the ITU-T Rec. P.910 please read:

ITU-T Recommendation P.910, Subjective video quality assessment methods for multimedia applications. Geneva: International Telecommunication Union, 2021.

Technical description of the implementation and validation is given in this paper:

• A crowdsourcing approach to video quality assessment Babak Naderi, Ross Cutler, arXiv preprint arXiv:2204.06784 (2022).

Citation

If you use this tool in your research please cite it with the following references:

@article{naderi2024,
  title={A crowdsourcing approach to video quality assessment},
  author={Naderi, Babak and Cutler, Ross},
  booktitle={ICASSP},
  year={2024}
}

Getting Started

• Preparation
• Running the Crowdsourcing Test
• Analyzing Data

Troubleshooting

For bug reports and issues with this code, please see the github issues page. Please review this page before contacting the authors.

Contact

Contact Ross Cutler with any questions.

License

Code License

MIT License

Copyright (c) Microsoft Corporation.

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Contributing

This project welcomes contributions and suggestions. Most contributions require you to agree to a Contributor License Agreement (CLA) declaring that you have the right to, and actually do, grant us the rights to use your contribution. For details, visit https://cla.opensource.microsoft.com.

When you submit a pull request, a CLA bot will automatically determine whether you need to provide a CLA and decorate the PR appropriately (e.g., status check, comment). Simply follow the instructions provided by the bot. You will only need to do this once across all repos using our CLA.

This project has adopted the Microsoft Open Source Code of Conduct. For more information see the Code of Conduct FAQ or contact [email protected] with any additional questions or comments.

Trademarks

This project may contain trademarks or logos for projects, products, or services. Authorized use of Microsoft trademarks or logos is subject to and must follow Microsoft's Trademark & Brand Guidelines. Use of Microsoft trademarks or logos in modified versions of this project must not cause confusion or imply Microsoft sponsorship. Any use of third-party trademarks or logos are subject to those third-party's policies.

p.910's Issues

Add Report Problem button

Allows raters to include a description of the problem and also include:

1. the video URLs being viewed
2. the browser version info

Include expected times in the instructions

To better allow the raters to gauge the HIT effort include the expected time, e.g., 5 min training, 5 min per HIT after.

Validate this is achieved using results.

Include loading % status not just a spinning circle

The spinning circle is not informative or functional. Include % instead.

create_trapping_clips: "arial.ttf" not found on Linux

"arial.ttf" cannot be found on Ubuntu. Could maybe add an optional input argument for the font?

Start creating files
Traceback (most recent call last):
  File "create_trapping_clips.py", line 207, in <module>
    n_created_files = create_trap_db(tp_cfg, args.source, args.des)
  File "create_trapping_clips.py", line 129, in create_trap_db
    msg_videos = create_msg_video(cfg, des, s_f)
  File "create_trapping_clips.py", line 44, in create_msg_video
    create_msg_img(cfg, score, des, width, height)
  File "create_trapping_clips.py", line 85, in create_msg_img
    font = ImageFont.truetype("arial.ttf", font_size)
  File "/home/gmittag/miniconda3/lib/python3.8/site-packages/PIL/ImageFont.py", line 959, in truetype
    return freetype(font)
  File "/home/gmittag/miniconda3/lib/python3.8/site-packages/PIL/ImageFont.py", line 956, in freetype
    return FreeTypeFont(font, size, index, encoding, layout_engine)
  File "/home/gmittag/miniconda3/lib/python3.8/site-packages/PIL/ImageFont.py", line 247, in __init__
    self.font = core.getfont(
OSError: cannot open resource

Add sufficient comments to code

The code is largely uncommented. These are required comments:

1. Describe what each function does
2. Describe the input and output of each function
3. Any non-trivial code should have at least a one-line comment describing what it will do

The HitApp server has no way to remove old projects

Improve reject messages

The Reject message "Make sure you follow the instruction: Setup section was unsuccessful." is not descriptive enough - please list what failed.

Also how can raters get past the setup stage? They should not be allowed to.

If rater has >= N accept and use failures then stop that rater from continuing

This prevents raters who are giving random answers from wasting their and our time.

Add these features to results_parser.py

• Don't use HITS with raters with < X% accuracy and >= N hits, e.g., X=80%, N=5
• Create a list of low-performance raters to ban

Update documentation in https://github.com/microsoft/P.910/blob/main/docs/results.md.

Block_Matrix_url column values do not fully match with available images

Preparation process No. 4 (optional): (optional) Upload the general resources (found in src\template\assets\imgs) in a cloud server and change the URLs associated to them as described in [General Resources](https://github.com/microsoft/P.910/blob/main/docs/general_res.md) asks us to upload a set of available images, and then match corresponding urls to the corresponding block_matrix_url column value, however, for some of the values, there is no match to an available resource.

The color blind image order should be be randomized

Unable to view test before publishing AMT batch

There is an error when accessing the HTML generated by the HIT App server, after it has been uploaded as a survey on MTurk with the data CSV. The loading status gets stuck at loading the videos (which are publicly accessible). This results in an error claiming internet speed as being the cause of the test not loading. This is even after setting min_average_download_speed_Mbps: 1 in master.cfg for that particular survey. This makes it difficult to debug what the underlying problem actually is.

Add CCR method

For creating an FRVQA tool CCR is needed, as it should make the network much simpler with a reference and tests show that ML codecs can improve not just degrade subjective quality.

Provide more diagnostic info if rater fails qualification

The required internet speed and monitor resolution should be in the HIT.

If a rater doesn't have the required internet speed or monitor resolution this should be specifically noted. I get questions about raters not knowing why they are qualified.