Annotate better with CVAT, the industry-leading data engine for machine learning. Used and trusted by teams at any scale, for data of any scale.
MIT License
Bot releases are visible (Hide)
Published by nmanovic about 2 years ago
Human Pose, Facial Landmarks, and similar use cases. See annotation with skeletons. Import and export of COCO Keypoints format.
Frame deleting in a job
Better tags visualization in classification mode
Serverless functions can return objects with attributes
CVAT SDK on PyPI. Install using pip install cvat-sdk
CVAT CLI on PyPI. Install using pip install cvat-cli
Improved helm chart
Full Changelog: https://github.com/opencv/cvat/compare/v2.1.0...v2.2.0
Published by nmanovic over 2 years ago
Published by nmanovic over 2 years ago
Permissions management system using open policy agent (https://www.openpolicyagent.org/). Now it is possible to create organizations and work on projects, tasks, and jobs in a team.
Rotated bounding boxes (https://openvinotoolkit.github.io/cvat/v2.0.0/docs/manual/advanced/annotation-with-rectangles/#rotation-rectangle)
Ellipses (https://openvinotoolkit.github.io/cvat/v2.0.0/docs/manual/advanced/annotation-with-ellipses/)
Jobs page with filtration capabilities (https://openvinotoolkit.github.io/cvat/v2.0.0/docs/manual/basics/jobs-page/)
Import datasets in public formats like MS COCO, Pascal VOC, YOLO into a project (https://openvinotoolkit.github.io/cvat/v2.0.0/docs/manual/advanced/export-import-datasets/#import-dataset)
Google Cloud Storage support in UI (https://openvinotoolkit.github.io/cvat/v2.0.0/docs/manual/basics/attach-cloud-storage/#using-google-cloud-storage).
OpenCV.js TrackerMIL tracker (https://openvinotoolkit.github.io/cvat/v2.0.0/docs/manual/advanced/opencv-tools/#trackermil)
Import and export new dataset formats: KITTI, LFW, Cityscapes, Open Images v6 (https://openvinotoolkit.github.io/cvat/v2.0.0/docs/manual/advanced/formats/)
YOLOv5 serverless function for automatic annotation. Possibility to automatically annotate tags and attributes using serverless functions (https://openvinotoolkit.github.io/cvat/v2.0.0/docs/manual/advanced/serverless-tutorial/)
Player option: Smooth image when zoom-in (left: smoothing, right: without smoothing)
Options to change font size & position of text labels on the canvas
API versioning scheme: accept header versioning instead of namespace versioning. The new version of REST API is v2.0.0-alpha. We are not going to support backward compatibility for now. Please expect that it will be changed even without its version update.
GET /api/jobs/<id>/commits
was implemented (https://github.com/openvinotoolkit/cvat/pull/4368)Published by nmanovic almost 3 years ago
Added OpenVINO logo and links to OpenVINO toolkit to the CVAT login page
Implemented advanced identity access management system (IAM) using Open Policy Agent. Now CVAT has configurable rules to provide access to different resources on the server. Users can create organizations to manage resources together. We had more than 15 issues on GitHub which described different use cases from community and Intel customers (e.g., #3950, #2708, #2637, #2681).
Implemented rotated bounding boxes. The feature was requested by our community multiple times to annotate objects in different scenarios.
Added new annotation formats for import/export of datasets: KITTI, LFW, Cityscapes, Open Images v6
Implemented backup of projects to move them from one CVAT instance to another easily (previously it was implemented only for annotation tasks):
Implemented import of datasets. Now it is possible to create a project from datasets in MS COCO, Pascal VOC, ImageNet and other formats.
Published by nmanovic almost 3 years ago
DataCloneError: The object could not be cloned
(https://github.com/openvinotoolkit/cvat/pull/3733)dataset_repo
application (https://github.com/openvinotoolkit/cvat/pull/3827)Published by nmanovic about 3 years ago
Read about these and other features in documentation: https://openvinotoolkit.github.io/cvat/docs/
Published by nmanovic about 3 years ago
Read about these and other features in documentation: https://openvinotoolkit.github.io/cvat/docs/
is_active
for user list (https://github.com/openvinotoolkit/cvat/pull/3235)docker-compose.yml
(https://github.com/openvinotoolkit/cvat/pull/3409)data/meta
requests (https://github.com/openvinotoolkit/cvat/pull/3269)cvat_ui
image build in case of OOM error that leads to the default nginx welcome pagePublished by nmanovic over 3 years ago
Published by nmanovic over 3 years ago
Published by nmanovic almost 4 years ago
next
query parameter (https://github.com/openvinotoolkit/cvat/pull/2527)docker-compose
file version from 2.3
to 3.3
(https://github.com/openvinotoolkit/cvat/pull/2235)outside
(https://github.com/openvinotoolkit/cvat/pull/2198)npm ci
instead of npm install
(https://github.com/openvinotoolkit/cvat/pull/2350)umask 077
in .bashrc file (https://github.com/openvinotoolkit/cvat/pull/2293)outside
in CVAT for Images
export (https://github.com/openvinotoolkit/cvat/pull/2345)jpg
or png
images in TF Detection API format (https://github.com/openvinotoolkit/datumaro/issues/35)Published by nmanovic almost 4 years ago
next
query parameter (https://github.com/openvinotoolkit/cvat/pull/2527)Published by nmanovic almost 4 years ago
docker-compose
file version from 2.3
to 3.3
(https://github.com/openvinotoolkit/cvat/pull/2235)outside
(https://github.com/openvinotoolkit/cvat/pull/2198)npm ci
instead of npm install
(https://github.com/openvinotoolkit/cvat/pull/2350)umask 077
in .bashrc file (https://github.com/openvinotoolkit/cvat/pull/2293)outside
in CVAT for Images
export (https://github.com/openvinotoolkit/cvat/pull/2345)jpg
or png
images in TF Detection API format (https://github.com/openvinotoolkit/datumaro/issues/35)Published by nmanovic about 4 years ago
stats
command, which shows some dataset statistics like image mean and std (https://github.com/opencv/cvat/pull/1734)convert
command to convert datasets directly (https://github.com/opencv/cvat/pull/1837)COCO
format does not convert bboxes to polygons on export (https://github.com/opencv/cvat/pull/1953)outside
annotations should not be in exported images (https://github.com/opencv/cvat/issues/1620)CVAT for video format
import error with interpolation (https://github.com/opencv/cvat/issues/1893)Image compression
definition mismatch (https://github.com/opencv/cvat/issues/1900)keyframe = True
(https://github.com/opencv/cvat/pull/1937)tensorflow
with no AVX instructions available (https://github.com/opencv/cvat/pull/1567)CVE-2020-9402
(https://github.com/opencv/cvat/pull/1657)Published by nmanovic about 4 years ago
COCO
format does not convert bboxes to polygons on export (https://github.com/opencv/cvat/pull/1953)outside
annotations should not be in exported images (https://github.com/opencv/cvat/issues/1620)CVAT for video format
import error with interpolation (https://github.com/opencv/cvat/issues/1893)Image compression
definition mismatch (https://github.com/opencv/cvat/issues/1900)keyframe = True
(https://github.com/opencv/cvat/pull/1937)Published by nmanovic over 4 years ago
stats
command, which shows some dataset statistics like image mean and std (https://github.com/opencv/cvat/pull/1734)tensorflow
with no AVX instructions available (https://github.com/opencv/cvat/pull/1567)CVE-2020-9402
(https://github.com/opencv/cvat/pull/1657)Published by nmanovic over 4 years ago
datumaro_project
export format (https://github.com/opencv/cvat/pull/1352)GET /task/<id>/annotations
: parameters are format
, filename
(now optional), action
(optional) (https://github.com/opencv/cvat/pull/1352)dataset/formats
, changed format of annotation/formats
(https://github.com/opencv/cvat/pull/1352)annotation
application is replaced with dataset_manager
(https://github.com/opencv/cvat/pull/1352)_DATUMARO_INIT_LOGLEVEL
env. variable is removed in favor of regular --loglevel
cli parameter (https://github.com/opencv/cvat/pull/1583)extract
command (https://github.com/opencv/cvat/pull/1352)cvat
export format in Datumaro (https://github.com/opencv/cvat/issues/1494)tf.placeholder() is not compatible with eager execution
exception for auto_segmentation (https://github.com/opencv/cvat/pull/1562)z_order
having no effect on segmentations (https://github.com/opencv/cvat/pull/1589)Published by nmanovic over 4 years ago
import
and export
to import/export raw annotations for Job and Task in cvat-core
(https://github.com/opencv/cvat/pull/1406)cvat-core
, cvat-canvas
, cvat-ui
). Initial versions are set to 1.0.0 (https://github.com/opencv/cvat/pull/1448)Published by nmanovic over 4 years ago
__undefined__
(invisibility, no shortcuts to be set in AAM)esc
when drawing instead of cancellationFileNotFoundError
during dump after moving format filesPublished by nmanovic over 4 years ago
Published by nmanovic over 4 years ago
occluded
and z_order
annotation attributes are now correctly passed to Datumaro (#1271)frame_XXXXXX[.ext]
format (#1274)