Fboylu (#551)

* Syncing staging <> master branches (#476)

* Removed submodules.

* Add back submodules using https://

* simplified plotting functions

* fixed most tests

* fixed test

* fixed unit test

* small text edits to the 02 notebook

* added fct description

* tiny cleanup on notebook

* Add pretrained keypoint model (#453)

* Add pretrained keypoint model

* Fix bugs in tests

* Add 03 notebook in conftest.py

* Minor revision

* Reformat code using black

* if folder exists, remove (#448)

* Add mask annotation tool (#447)

* Add mask annotation tool

* Update mask annotation explanation and add converion scripts

* Add screenshots of Labelbox annotation

* Rearrange screenshots

* Move convertion script into functions in data.py

* Point out annotation conversion scripts clearly in notebook

* Refine annotation conversion scripts

* Fix bugs

* Add tests for labelbox format conversion methods

* Move r2p1d from contrib to scenarios.

* Update .gitignore.

* Add README.md

* Remove the folder /scenario/action_recognition/data/samples; update notebook to use web url for sample data.

* Move data split files to data/misc; update notebook accordingly.

* Update data path.

* Add keypoint detection with tuned model (#454)

* Add keypoint detetion with tuned model

* Add tests

* Minor revision

* Update tests

* Fix bugs in tests

* Use GPU device if available

* Update tests

* Fix bug: 'not idx' will be 'True' if 'idx=0'

* Fix bugs

* Move toy keypoint meta into notebook

* Fix bugs

* Fix bugs

* Fix bugs in notebook

* Add descriptions for keypoint meta data

* Raise exception when RandomHorizontalFlip is used without specifying hflip_inds

* Add NOTICE file.

* Add keypoint detection model tuning with top and bottom keypoints (#456)

* Add keypoint detection model tuning with top and bottom keypoints

* Fix undefined unzip_url

* Resolved undefined od_urls

* Add annotation tool to scenarios.

* Plot keypoints as round dots to make them noticeable (#458)

* Plot keypoints as dots

* Change variable naming

* Resolve test machine failure (#460)

This is due to the latest PyTorch (version 1.3) from conda is built on
CUDA 10.1 while the version on the test machine is CUDA 10.0.

* Remove unused imports in 02_mask_rcnn.ipynb (#463)

* Remove unused imports in 02_mask_rcnn.ipynb

* Add missing imports

* Simplify binary_mask() (#464)

* clean up of keypoint detection notebook

* minor clean-up

* remove conflict code (#471)

* Update README.md (#472)

* updated readmes

* added images

* updated readmes

* added intro figure

* modified new picture

* update figure

* Move to bare Windows GPU VMs and fix build issues (#475)

* Updating AzureDevOps pipeline agent pool to use a barebones Windows GPU VM.

* Added a AzureDevOps pipeline for windows.

* Added a conda init step for powershell.

* Adding conda to PATH as a separate step

* Pinning pillow to 6.1 to fix issues listed in: https://github.com/python-pillow/Pillow/issues/4130

* chained powershell commands

* Updated AzureDevOps yml file to use the inline powershell script syntax.

* Moved to using inline commands + Invoke-Expression syntax

* Adding indentation for chained commands.

* Moving to the literal block style indicator for yaml for chained commands.

* Using call activate instead of source activate when activating the conda environment through powershell.

* Switching to script steps instead of using powershell steps for conda operations.

* Fixed the numpy.float64 issue by moving to a fork with the upstream fixes for the issue.

* Fixed the numpy.float64 issue in utils/cv/detection/plot.py

Co-authored-by: PatrickBue <pabuehle@microsoft.com>
Co-authored-by: Simon Zhao <43029286+simonzhaoms@users.noreply.github.com>
Co-authored-by: Miguel González-Fierro <3491412+miguelgfierro@users.noreply.github.com>
Co-authored-by: JS <jiata@microsoft.com>

* fix to load from saved mask-rcnn model

* fix to load from saved mask-rcnn model

* fix to load from saved mask-rcnn model

Co-authored-by: Young Park <youngpark@cs.stanford.edu>
Co-authored-by: PatrickBue <pabuehle@microsoft.com>
Co-authored-by: Simon Zhao <43029286+simonzhaoms@users.noreply.github.com>
Co-authored-by: Miguel González-Fierro <3491412+miguelgfierro@users.noreply.github.com>
Co-authored-by: JS <jiata@microsoft.com>

utils_cv/detection/model.py

@@ -803,7 +803,7 @@ class DetectionLearner:
             self.labels = meta_data["labels"]
 
     @classmethod
-    def from_saved_model(cls, name: str, path: str) -> "DetectionLearner":
+    def from_saved_model(cls, name: str, path: str, mask: bool = False) -> "DetectionLearner":
         """ Create an instance of the DetectionLearner from a saved model.
 
         This function expects the format that is outputted in the `save`
@@ -812,6 +812,7 @@ class DetectionLearner:
         Args:
             name: the name of the model you wish to load
             path: the path to get your model from
+            mask: if the model is an instance of maskrcnn
 
         Returns:
             A DetectionLearner object that can inference.
@@ -827,9 +828,15 @@ class DetectionLearner:
             im_size = meta_data["im_size"]
             labels = meta_data["labels"]
 
-        model = get_pretrained_fasterrcnn(
+        if mask:
+            model = get_pretrained_maskrcnn(
             len(labels) + 1, min_size=im_size, max_size=im_size
-        )
+            )
+        else:
+            model = get_pretrained_fasterrcnn(
+            len(labels) + 1, min_size=im_size, max_size=im_size
+            )
+
         detection_learner = DetectionLearner(model=model, labels=labels)
         detection_learner.load(name=name, path=path)
         return detection_learner