cvat

Changed

• Do not reload annotation view when renew the job or update job state (https://github.com/opencv/cvat/pull/6851)
• Now images from cloud buckets are loaded in parallel when preparing a chunk (https://github.com/opencv/cvat/pull/6881)

Fixed

• Downloading additional data from cloud storage if use_cache=true and job_file_mapping are specified
  (https://github.com/opencv/cvat/pull/6879)
• Leaving an organization (https://github.com/opencv/cvat/pull/6422)

Full Changelog: https://github.com/opencv/cvat/compare/v2.7.1...v2.7.2

[2.7.1] - 2023-09-15

Fixed

• Include cloud storage manifest file to selected files if manifest was used as data source (https://github.com/opencv/cvat/pull/6850)
• Keep sequence of files when directories were specified in server_files (https://github.com/opencv/cvat/pull/6850)

Full Changelog: https://github.com/opencv/cvat/compare/v2.7.0...v2.7.1

Added

• Admin actions for easy activation/deactivation of users (https://github.com/opencv/cvat/pull/6314)

Fixed

• Invalid input validation in for cloud_storage_id (https://github.com/opencv/cvat/pull/6825)
• Incorrect task progress report for 3rdparty users (https://github.com/opencv/cvat/pull/6834)

Security

• Security upgrade gitpython from 3.1.33 to 3.1.35 (https://github.com/opencv/cvat/pull/6843)
• Security upgrade numpy from 1.22.0 to 1.22.4 (https://github.com/opencv/cvat/pull/6843)

Full Changelog: https://github.com/opencv/cvat/compare/v2.6.2...v2.7.0

Added

• Gamma correcton filter (https://github.com/opencv/cvat/pull/6771)
• Introduced the feature to hide or show objects in review mode (https://github.com/opencv/cvat/pull/6808)

Changed

• [Helm] Database migrations are now executed as a separate job,
  rather than in the server pod, to mitigate the risk of data
  corruption when using multiple server replicas
  (https://github.com/opencv/cvat/pull/6780)
• Clicking multiple times on icons in the left
  sidebar now toggles the corresponding popovers open and closed
  (https://github.com/opencv/cvat/pull/6817)
• Transitioned to using KeyDB with FLASH for data
  chunk caching, replacing diskcache (https://github.com/opencv/cvat/pull/6773)

Removed

• Removed outdated use of hostnames when accessing Git, OpenCV, or analytics via the UI (https://github.com/opencv/cvat/pull/6799)
• Removed the Feedback/Share component (https://github.com/opencv/cvat/pull/6805)

Fixed

• Resolved the issue of the canvas zooming while scrolling
  through the comments list in an issue (https://github.com/opencv/cvat/pull/6758)
• Addressed the bug that allowed for multiple issue
  creations upon initial submission (https://github.com/opencv/cvat/pull/6758)
• Fixed the issue of running deep learning models on
  non-JPEG compressed TIFF images (https://github.com/opencv/cvat/pull/6789)
• Adjusted padding on the tasks, projects, and models pages (https://github.com/opencv/cvat/pull/6778)
• Corrected hotkey handlers to avoid overriding default behavior when modal windows are open
  (https://github.com/opencv/cvat/pull/6800)
• Resolved the need to move the mouse to activate
  brush or eraser effects; a single click is now sufficient (https://github.com/opencv/cvat/pull/6800)
• Fixed a memory leak issue in the logging system (https://github.com/opencv/cvat/pull/6804)
• Addressed a race condition that occurred during the initial creation of secret_key.py
  (https://github.com/opencv/cvat/pull/6775)
• Eliminated duplicate log entries generated by the CVAT server
  (https://github.com/opencv/cvat/pull/6766)

Full Changelog: https://github.com/opencv/cvat/compare/v2.6.1...v2.6.2

Added

• More information about task progress on tasks page (https://github.com/opencv/cvat/pull/5723)
• Prefetching next chunk when user navigates by frames manually (https://github.com/opencv/cvat/pull/6695)

Changed

• Bumped nuclio version to 1.11.24 and removed /tmp mounting in the nuclio container to adhere the update.
• Response code for empty cloud storage preview 204 -> 404 (https://github.com/opencv/cvat/pull/6727)
• Organization now opened immediately after it is created (https://github.com/opencv/cvat/pull/6705)
• More responsive automatic annotation progress bar (https://github.com/opencv/cvat/pull/6734)
• Improved message when invite more users to an organization (https://github.com/opencv/cvat/pull/6731)

Fixed

• Exporting project when its tasks has not data (https://github.com/opencv/cvat/pull/6658)
• Removing job assignee (https://github.com/opencv/cvat/pull/6712)
• UI fail when select a mask or a skeleton with center-aligned text (https://github.com/opencv/cvat/pull/6753)
• Fixed switching from organization to sandbox while getting a resource (https://github.com/opencv/cvat/pull/6689)
• You do not have permissions when user is cancelling automatic annotation (https://github.com/opencv/cvat/pull/6734)
• Automatic annotation progress bar is invisible if the app initialized on the task page
  (https://github.com/opencv/cvat/pull/6734)
• Extra status check requests for automatic annotation (https://github.com/opencv/cvat/pull/6734)
• [SDK]: FileExistsError exception raised on Windows when a dataset is loaded from cache
  (https://github.com/opencv/cvat/pull/6722)

Security

• Remote Code Execution (RCE) SNYK-PYTHON-GITPYTHON-5840584

Added

• [SDK] Introduced the DeferredTqdmProgressReporter class, which avoids the glitchy output seen with the TqdmProgressReporter under certain circumstances (https://github.com/opencv/cvat/pull/6556)
• [SDK, CLI] Added the cvat_sdk.auto_annotation module, providing functionality to automatically annotate tasks by executing a user-provided function on the local machine. A corresponding CLI command (auto-annotate) is also available. Some predefined functions using torchvision are also available. (https://github.com/opencv/cvat/pull/6483, https://github.com/opencv/cvat/pull/6649)
• Included an indication for cached frames in the interface (https://github.com/opencv/cvat/pull/6586)

Changed

• Raised the default guide assets limitations to 30 assets, with a maximum size of 10MB each (https://github.com/opencv/cvat/pull/6575)
• [SDK] Custom ProgressReporter implementations should now override start2 instead of start The old implementation is still supported.
  (https://github.com/opencv/cvat/pull/6556)
• Improved memory optimization and code in the decoding module (https://github.com/opencv/cvat/pull/6585)

Removed

• Removed the YOLOv5 serverless function (https://github.com/opencv/cvat/pull/6618)

Fixed

• Corrected an issue where the prebuilt FFmpeg bundled in PyAV was being used instead of the custom build.
• Fixed the filename for labels in the CamVid format (https://github.com/opencv/cvat/pull/6600)

Added

• We've added support for multi-line text attributes (https://github.com/opencv/cvat/pull/6458)
• You can now set a default attribute value for SELECT, RADIO types on UI
  (https://github.com/opencv/cvat/pull/6474)
• [SDK] cvat_sdk.datasets, is now available, providing a framework-agnostic alternative to cvat_sdk.pytorch
  (https://github.com/opencv/cvat/pull/6428)
• We've introduced analytics for Jobs, Tasks, and Project (https://github.com/opencv/cvat/pull/6371)

Changed

• [Helm] In Helm, we've added a configurable default storage option to the chart (https://github.com/opencv/cvat/pull/6137)

Removed

• [Helm] In Helm, we've eliminated the obligatory use of hardcoded traefik ingress (https://github.com/opencv/cvat/pull/6137)

Fixed

• Fixed an issue with calculating the number of objects on the annotation view when frames are deleted
  (https://github.com/opencv/cvat/pull/6493)
• [SDK] In SDK, we've fixed the issue with creating attributes with blank default values
  (https://github.com/opencv/cvat/pull/6454)
• [SDK] We've corrected a problem in SDK where it was altering input data in models (https://github.com/opencv/cvat/pull/6455)
• Fixed exporting of hash for shapes and tags in a specific corner case (https://github.com/opencv/cvat/pull/6517)
• Resolved the issue where 3D jobs couldn't be opened in validation mode (https://github.com/opencv/cvat/pull/6507)
• Fixed SAM plugin (403 code for workers in organizations) (https://github.com/opencv/cvat/pull/6514)
• Fixed the issue where initial frame from query parameter was not opening specific frame in a job
  (https://github.com/opencv/cvat/pull/6506)
• Corrected the issue with the removal of the first keyframe (https://github.com/opencv/cvat/pull/6494)
• Fixed the display of project previews on small screens and updated stylelint & rules (https://github.com/opencv/cvat/pull/6551)
• Implemented server-side validation for attribute specifications
  (https://github.com/opencv/cvat/pull/6447)
• [API] Fixed API issue related to file downloading failures for filenames with special characters (https://github.com/opencv/cvat/pull/6492)
• [Helm] In Helm, we've resolved an issue with multiple caches
  in the same RWX volume, which was preventing db migration from starting (https://github.com/opencv/cvat/pull/6137)

Full Changelog: https://github.com/opencv/cvat/compare/v2.5.1...v2.5.2

[2.5.1] - 2023-07-19

Fixed

• Memory leak related to unclosed av container (https://github.com/opencv/cvat/pull/6501)

Added

• Now CVAT supports project/task markdown description with additional assets
  (png, jpeg, gif, webp images and pdf files) (https://github.com/opencv/cvat/pull/6191)
• Ground Truth jobs and quality analytics for tasks (https://github.com/opencv/cvat/pull/6039)

Fixed

• The problem with manifest file in tasks restored from backup (https://github.com/opencv/cvat/issues/5971)
• The problem with task mode in a task restored from backup (https://github.com/opencv/cvat/issues/5668)
• Visible 'To background' button in review mode (https://github.com/opencv/cvat/pull/6363)
• Added missed auto_add argument to Issue model (https://github.com/opencv/cvat/pull/6364)
• [API] Performance of several API endpoints (https://github.com/opencv/cvat/pull/6340)
• [API] Invalid schema for the owner field in several endpoints (https://github.com/opencv/cvat/pull/6343)
• Some internal errors occurring during lambda function invocations
  could be mistakenly reported as invalid requests
  (https://github.com/opencv/cvat/pull/6394)
• [SDK] Loading tasks that have been cached with the PyTorch adapter
  (https://github.com/opencv/cvat/issues/6047)
• The problem with importing annotations if dataset has extra dots in filenames
  (https://github.com/opencv/cvat/pull/6350)

Security

• More comprehensive SSRF mitigations were implemented.
  Previously, on task creation it was prohibited to specify remote data URLs
  with hosts that resolved to IP addresses in the private ranges.
  Now, redirects to such URLs are also prohibited.
  In addition, this restriction is now also applied to webhook URLs.
  System administrators can allow or deny custom IP address ranges
  with the SMOKESCREEN_OPTS environment variable.
  (https://github.com/opencv/cvat/pull/6362).

Full Changelog: https://github.com/opencv/cvat/compare/v2.4.9...v2.5.0

Fixed

• Error related to calling serverless functions on some image formats (https://github.com/opencv/cvat/pull/6384). A lambda function was called without mandatory argument: 'JPG'

[2.4.8] - 2023-06-22

Fixed

• Getting original chunks for items in specific cases (https://github.com/opencv/cvat/pull/6355)

Full Changelog: https://github.com/opencv/cvat/compare/v2.4.7...v2.4.8

Added

• [API] API Now supports the creation and removal of Ground Truth jobs. (https://github.com/opencv/cvat/pull/6204)
• [API] We've introduced task quality estimation endpoints. (https://github.com/opencv/cvat/pull/6204)
• [CLI] An option to select the organization. (https://github.com/opencv/cvat/pull/6317)

Fixed

• Issues with running serverless models for EXIF-rotated images. (https://github.com/opencv/cvat/pull/6275/)
• File uploading issues when using https configuration. (https://github.com/opencv/cvat/pull/6308)
• Dataset export error with outside property of tracks. (https://github.com/opencv/cvat/issues/5971)
• Broken logging in the TransT serverless function. (https://github.com/opencv/cvat/pull/6290)

Added

• [Server API] An option to supply custom file ordering for task data uploads (https://github.com/opencv/cvat/pull/5083)
• New option semi-auto is available as annotations source (https://github.com/opencv/cvat/pull/6263)

Changed

• Allowed to use dataset manifest for the predefined sorting method for task data (https://github.com/opencv/cvat/pull/5083)

Changed

• Replaced Apache mod_wsgi with Uvicorn ASGI server for backend use(https://github.com/opencv/cvat/pull/6195)

Fixed

• Incorrect location of temporary file during job annotation import.(https://github.com/opencv/cvat/pull/5909)
• Deletion of uploaded file along with annotations/backups when an RQ job
  has been initiated, but no subsequent status check requests have been made.(https://github.com/opencv/cvat/pull/5909)
• Deletion of uploaded files, including annotations and backups,
  after they have been uploaded to the server using the TUS protocol but before an RQ job has been initiated. (https://github.com/opencv/cvat/pull/5909)
• Simultaneous creation of tasks or projects with identical names from backups by multiple users.(https://github.com/opencv/cvat/pull/5909)
• [Server API] The predefined sorting method for task data uploads (https://github.com/opencv/cvat/pull/5083)
• Allowed slashes in export filenames. (https://github.com/opencv/cvat/pull/6265)

New Contributors

• @pktiuk made their first contribution in https://github.com/opencv/cvat/pull/6164
• @senthilkumarkj made their first contribution in https://github.com/opencv/cvat/pull/6163
• @pbarsic made their first contribution in https://github.com/opencv/cvat/pull/6249

Full Changelog: https://github.com/opencv/cvat/compare/v2.4.5...v2.4.6

[2.4.5] - 2023-06-02

Added

• Integrated support for sharepoint and cloud storage files, along with
  directories to be omitted during task creation (server) (https://github.com/opencv/cvat/pull/6074)
• Enabled task creation with directories from cloud storage or sharepoint (https://github.com/opencv/cvat/pull/6074)
• Enhanced task creation to support any data type supported by the server
  by default, from cloud storage without the necessity for the use_cache option (https://github.com/opencv/cvat/pull/6074)
• Added capability for task creation with data from cloud storage without the use_cache option (https://github.com/opencv/cvat/pull/6074)

Changed

• User can now access resource links from any organization or sandbox, granted it's available to them (https://github.com/opencv/cvat/pull/5892)
• Cloud storage manifest files have been made optional (https://github.com/opencv/cvat/pull/6074)
• Updated Django to the 4.2.x version (https://github.com/opencv/cvat/pull/6122)
• Renamed certain Nuclio functions to adhere to a common naming convention. For instance,
  onnx-yolov7 -> onnx-wongkinyiu-yolov7, ultralytics-yolov5 -> pth-ultralytics-yolov5
  (https://github.com/opencv/cvat/pull/6140)

Deprecated

• Deprecated the endpoint /cloudstorages/{id}/content (https://github.com/opencv/cvat/pull/6074)

Fixed

• Fixed the issue of skeletons dumping on created tasks/projects (https://github.com/opencv/cvat/pull/6157)
• Resolved an issue related to saving annotations for skeleton tracks (https://github.com/opencv/cvat/pull/6075)

Full Changelog: https://github.com/opencv/cvat/compare/v2.4.4...v2.4.5

Added

• Introduced a new configuration option for controlling the invocation of Nuclio functions.
  (https://github.com/opencv/cvat/pull/6146)

Changed

• Relocated SAM masks decoder to frontend operation.
  (https://github.com/opencv/cvat/pull/6019)
• Switched person-reidentification-retail-0300 and faster_rcnn_inception_v2_coco Nuclio functions with person-reidentification-retail-0277 and faster_rcnn_inception_resnet_v2_atrous_coco respectively.
  (https://github.com/opencv/cvat/pull/6129)
• Upgraded OpenVINO-based Nuclio functions to utilize the OpenVINO 2022.3 runtime.
  (https://github.com/opencv/cvat/pull/6129)

Fixed

• Resolved issues with tracking multiple objects (30 and more) using the TransT tracker.
  (https://github.com/opencv/cvat/pull/6073)
• Addressed azure.core.exceptions.ResourceExistsError: The specified blob already exists.
  (https://github.com/opencv/cvat/pull/6082)
• Corrected image scaling issues when transitioning between images of different resolutions.
  (https://github.com/opencv/cvat/pull/6081)
• Fixed inaccurate reporting of completed job counts.
  (https://github.com/opencv/cvat/issues/6098)
• Allowed OpenVINO-based Nuclio functions to be deployed to Kubernetes.
  (https://github.com/opencv/cvat/pull/6129)
• Improved skeleton size checks after drawing.
  (https://github.com/opencv/cvat/pull/6156)
• Fixed HRNet CPU serverless function.
  (https://github.com/opencv/cvat/pull/6150)
• Prevented sending of empty list of events.
  (https://github.com/opencv/cvat/pull/6154)

Full Changelog: https://github.com/opencv/cvat/compare/v2.4.3...v2.4.4

Changed

• Docker images no longer include Ubuntu package sources or FFmpeg/OpenH264 sources
  (https://github.com/opencv/cvat/pull/6040)
• TUS chunk size changed from 100 MB to 2 MB
  (https://github.com/opencv/cvat/pull/6058)

Full Changelog: https://github.com/opencv/cvat/compare/v2.4.2...v2.4.3

• Now SAM is available in CVAT self-hosted solution.
  Read more in The Facebook Segment Anything Model is now available in the self-hosted version of CVAT.
  

• Support for Azure Blob Storage connection string: This update adds support for Connection String authentication for Azure Blob Storage. This means that you can use Azure Blob Storage on the IoT Edge of an edge device.

• The issue related to webhook events not being sent has been resolved.

New Contributors

• @suzusuzu made their first contribution in https://github.com/opencv/cvat/pull/4649
• @Bassileios made their first contribution in https://github.com/opencv/cvat/pull/5731

Full Changelog: https://github.com/opencv/cvat/compare/v2.4.1...v2.4.2

Fixed

• Optimized annotations fetching up to 10 times (https://github.com/opencv/cvat/pull/5974)
• Incorrect calculation of working time in analytics (https://github.com/opencv/cvat/pull/5973)

In this release, we’ve introduced new features based on our vision and suggestions in the CVAT community. Thank you for reaching out and providing us with valuable feedback. Here is the list of features we are shipping in the CVAT v2.4.0

• Added the grid view and support of the contextual images for 2D and 3D tasks. For more information, see Contextual images.

  

• Added Interpolation for 3D cuboids. Also, annotation can be exported and imported in Datumaro and Supervisely Point Cloud formats.

• Added Yolo v7 serverless functionally for automated annotation.

• Collections in server API were redesigned for improved performance.

• Added simple filters for collection endpoints in API. For example, now you can filter the tasks by username: /tasks?project_id=42&user=myusername

• Improved analytics and enriched it with the following features: log collection from UI and server, exceptions collection, user activity visualization, aggregation of user and job working time, and log filtering for debugging purposes. Analytics are now based on ClickHouse + Vector + Grafana.

  

• Improved the work of the Pytorch adapter also known as the cvat_sdk.pytorch module, which allows you to use a CVAT project or task directly as a PyTorch-compatible dataset, eliminating the need for intermediate representation. For more information, see CVAT SDK Pytorch Adapter.

• Improved high-level SDK APIs for task data uploading. The default data source is now the local file system. Added an option to wait for data processing to finish after data uploading, the option is turned on by default.

• Added high-level API for working with organizations. For more information, see High-level API.

New Contributors

• @scottykwok made their first contribution in https://github.com/opencv/cvat/pull/5370
• @Bytestorm5 made their first contribution in https://github.com/opencv/cvat/pull/5558
• @hardikdava made their first contribution in https://github.com/opencv/cvat/pull/5552
• @michael-selasi-dzamesi made their first contribution in https://github.com/opencv/cvat/pull/5360
• @waqarsqureshi made their first contribution in https://github.com/opencv/cvat/pull/4665
• @MyNameIsFu made their first contribution in https://github.com/opencv/cvat/pull/5005
• @mats16 made their first contribution in https://github.com/opencv/cvat/pull/5533
• @vainkop made their first contribution in https://github.com/opencv/cvat/pull/5585
• @ddx-day made their first contribution in https://github.com/opencv/cvat/pull/5059
• @Sothatsit made their first contribution in https://github.com/opencv/cvat/pull/5609
• @saifkhichi96 made their first contribution in https://github.com/opencv/cvat/pull/5805
• @FObermaier made their first contribution in https://github.com/opencv/cvat/pull/5835
• @RyanHir made their first contribution in https://github.com/opencv/cvat/pull/5848

Full Changelog: https://github.com/opencv/cvat/compare/v2.3.0...v2.4.0

In this release, we’ve introduced new features based on our vision and suggestions in the CVAT community, plus addressed more than 20+ reported bugs. Thank you for reaching out and providing us with valuable feedback. Here is the list of features we are shipping:

• Added the Brush tool to create pixel masks for disjoint objects with multiple parts: a house hiding behind trees, a car behind a pedestrian, or a pillar behind a traffic sign. Read more.

  

• Added Webhooks for Projects and Organizations. Configure them to handle application notifications about changes in a specific project or organization. Read more.

  

• Added sorting by Z-Order: farthest from the camera objects go on the top of the list. Now sorted objects have an additional indication on the objects sidebar.

  

• Enabled leveraging of NVIDIA GPU for YOLO v5 serverless function. Now you can annotate with YOLO faster.

• Added social accounts authentication. Now you can log in to CVAT using your social auth accounts:

  • For App CVAT: Google and GitHub.
  • For local version: Google, GitHub, and the possibility to configure Social Auth with Django services.

• Added email authentication. Now you can authenticate in CVAT with your email instead of login.

  

• Added Propagation backward to the Confirm Propagation widget, now propagation works both ways.

  

New Contributors

• @AlexeyAlexeevXperienceAI made their first contribution in https://github.com/opencv/cvat/pull/4824
• @rhee-elten made their first contribution in https://github.com/opencv/cvat/pull/4947
• @KoheiKanagu made their first contribution in https://github.com/opencv/cvat/pull/4659
• @arvfilippov made their first contribution in https://github.com/opencv/cvat/pull/5014
• @saurabheights made their first contribution in https://github.com/opencv/cvat/pull/5030
• @tdhooghe made their first contribution in https://github.com/opencv/cvat/pull/4960
• @alex4men made their first contribution in https://github.com/opencv/cvat/pull/5101
• @inpv made their first contribution in https://github.com/opencv/cvat/pull/5192
• @ftconan made their first contribution in https://github.com/opencv/cvat/pull/4647
• @arkravchuk made their first contribution in https://github.com/opencv/cvat/pull/4712
• @mdacoca made their first contribution in https://github.com/opencv/cvat/pull/5269

Full Changelog: https://github.com/opencv/cvat/compare/v2.2.0...v2.3.0
CHANGELOG.md: https://github.com/opencv/cvat/blob/v2.3.0/CHANGELOG.md