tensorflow / tensorflow/models
Downloads of object detection models blocked by default on Chrome
@pkulzc is already working on this.
Since Oct 8, 2020.
- Dominant language
- Python
- Stars
- 77.7k
- Forks
- 44.8k
- PR merge metrics
- No merged PRs in 30d
Description
Prerequisites
Please answer the following question for yourself before submitting an issue.
- I checked to make sure that this issue has not been filed already.
1. The entire URL of the documentation with the issue
2. Describe the issue
Chrome is automatically blocking downloads of object detection models in the tf2 object detection zoo (though the block can be worked around by explicitly telling Chrome you want to download the file despite the security implications).
Since Chrome 84, Chrome blocks downloads by default if the download is over HTTP while the page the user is currently on is served over HTTPS (https://www.zdnet.com/article/google-to-block-some-http-file-downloads-starting-with-chrome-83/ , https://blog.chromium.org/2020/02/protecting-users-from-insecure.html). This is the case for the model zoo, where the models download over HTTP while Github is delivered over HTTPS.
Below is an example of a model download link that does not work when clicked from inside the model zoo (or when clicked from this issue):
http://download.tensorflow.org/models/object_detection/tf2/20200711/efficientdet_d0_coco17_tpu-32.tar.gz
I believe fixing this issue will be more involved than simply adding an s to make https as that will result in a cert domain name mismatch.
I would be glad to implement a fix for this issue, but I am not sure of the best course.

3. Steps to reproduce
- View the model zoo at https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf2_detection_zoo.md
- Try to download a model by clicking its name
- Observe that the download is blocked

Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Assessment
This issue has not been assessed yet.