This is pretty much the same fix applied to f-BRS in #5384. We've been
using HRNet for a while and now an then we receive "500 errors" just as
reported in #5299 when someone forgets to drop the alpha-channel from
our images.
### Motivation and context
The RuntimeError is a little bit different, but comes from the same
issue: RGBA instead of RGB images:
> RuntimeError: Given groups=1, weight of size [16, 3, 1, 1], expected
input[*, 4, *, *] to have 3 channels, but got 4 channels instead.
### How has this been tested?
I created a task with images with and w/o alpha channel, and the
interactor works on both now.
* Update curl command for Google Drive weights and adjust yaml files (#4736)
* update changelog
Co-authored-by: Boris Sekachev <boris.sekachev@yandex.ru>
* Added HRNet18 click based interactive segmentation (#3729)
The commit adds serverless container that has a click based
interactive segmentation model(HRNet18) that performs better than
f-BRS method statistically.
* fixup! Added HRNet18 click based interactive segmentation (#3729)
Co-authored-by: Nikita Manovich <nikita.manovich@intel.com>
* Improved removable points, added a button to finish current object
* Code refactoring for ai tools and opencv
* Fixed style for workspace selector
* IOG UX fix
* Updated version & changelog
* Added 'min_neg_points' parameter to serverless interactors
* Return 'min_neg_points' from the server
* Initial version of Inside Outside Guidance serverless function
* Support neg_points in interactors
* Improved deployment process of serverless functions
* Improve installation.md for serverless functions.
* Update CHANGELOG, use NUCLIO_DASHBOARD_DEFAULT_FUNCTION_MOUNT_MODE as
recommended by nuclio developers.
* Disable warning from markdown linter about max line length for a table.
* Fix IOG function with conda environment
* Fix tensorflow matterport/mask_rcnn
Co-authored-by: Boris Sekachev <boris.sekachev@intel.com>