You cannot select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
8bf647b360
* Release 0.5 (#705) * Changed version number (0, 5, 'final', 0). * Updated changelog file. * fixed default attribute values for tracked shapes (#703) * typo ? Should not this be cvat_redis -> redis ? * Fixed labels regex for non-latin characters (#708) * Update README.md * Update README.md * Don't save shapes with keyframe==False * Selecting non images leads to 400 error (#734) * Fix HTTP 400 error if together with vision data the user submit non-vision data (e.g. text files) * Ignore SVG images because Pillow doesn't work with them. * Fix the problem with duplicated frames in case of "share" (#735) * Fix the problem with duplicated frames in case of "share". * Fix a case when the code works incorrectly /a/b/c /a/b/c0 Previously only /a/b/c will be in output but should be both. * added method docs to Auto Annotation inference.py (#725) * remove deprecated method call `from_ir` (#726) * New command line tool for working with tasks (#732) * Adding new command line tool for performing common task related operations (create, list, delete, etc.) * Replaced @exception decorator with try/except in main() * Replaced optional --name with positional name and removed default * Added license text to files * Added django units to cover future API changes * Refactored into submodules to better support tests * Fix an issue with permissions (observer can change annotations) (#745) * Fixed a problem with observer (check_object_permissions method was not called) * Added a test case to cover issue #712. * COCO Annotation IDs should begin with 1 (#748) Currently the annotation ID begins with 0 which is interpreted by cocoapi as a false detection. The array dtm saves the matches via the ground truth annotation ID. The variable dtm is initialized as an array of zeros. |
6 years ago | |
|---|---|---|
| .. | ||
| README.md | 6 years ago | |
| docker-compose.auto_segmentation.yml | 6 years ago | |
| install.sh | 6 years ago | |
README.md
Keras+Tensorflow Mask R-CNN Segmentation
What is it?
-
This application allows you automatically to segment many various objects on images.
-
It's based on Feature Pyramid Network (FPN) and a ResNet101 backbone.
-
It uses a pre-trained model on MS COCO dataset
-
It supports next classes (use them in "labels" row):
'BG', 'person', 'bicycle', 'car', 'motorcycle', 'airplane',
'bus', 'train', 'truck', 'boat', 'traffic light',
'fire hydrant', 'stop sign', 'parking meter', 'bench', 'bird',
'cat', 'dog', 'horse', 'sheep', 'cow', 'elephant', 'bear',
'zebra', 'giraffe', 'backpack', 'umbrella', 'handbag', 'tie',
'suitcase', 'frisbee', 'skis', 'snowboard', 'sports ball',
'kite', 'baseball bat', 'baseball glove', 'skateboard',
'surfboard', 'tennis racket', 'bottle', 'wine glass', 'cup',
'fork', 'knife', 'spoon', 'bowl', 'banana', 'apple',
'sandwich', 'orange', 'broccoli', 'carrot', 'hot dog', 'pizza',
'donut', 'cake', 'chair', 'couch', 'potted plant', 'bed',
'dining table', 'toilet', 'tv', 'laptop', 'mouse', 'remote',
'keyboard', 'cell phone', 'microwave', 'oven', 'toaster',
'sink', 'refrigerator', 'book', 'clock', 'vase', 'scissors',
'teddy bear', 'hair drier', 'toothbrush'.
- Component adds "Run Auto Segmentation" button into dashboard.
Build docker image
# From project root directory
docker-compose -f docker-compose.yml -f components/auto_segmentation/docker-compose.auto_segmentation.yml build
Run docker container
# From project root directory
docker-compose -f docker-compose.yml -f components/auto_segmentation/docker-compose.auto_segmentation.yml up -d