Release 7.8
New Feature | Improvement | Bug Fix | Enterprise Only |
---|---|---|---|
Model
Status | Details |
---|---|
Model: Published "person-detection-edge" Visual-Detector | |
Model: Published "person-vehicle-detection-edge" Visual Detector |
Platform-Object
Status | Details |
---|---|
Annotation: Added support for Time-Segments | |
Application: Reindexing with no inputs returns invalid request | |
Model-Type: Added Track-Aggregator | |
Model: Added Model.Notes field | |
Task: Continuously check Consensus threshold logic | |
Task: Fix incorrect status mapping in Annotation stats | |
Task: Ignore Consensus logic on "Empty" annotations |
Portal
Status | Details |
---|---|
Input-Details: Audio doesn't stop when exiting after playing video | |
Task-Labeler: Added histogram showing count of objects throughout Video-Inputs | |
Task-Labeler: Added Time-Segment tool | |
Task-Labeler: Added tracklet visualization for Object-Track labels | |
Task-Labeler: AI-Assist suggestions not displaying for Public Visual-Classifiers | |
Task-Labeler: Fix carousel being pushed off-screen by timeline | |
Task-Labeler: Loading animation continues indefinitely after GetTaskInputs returns no hits | |
Task-Manager: Reduce minimum Consensus Threshold to 2 | |
Task-Reviewer: Video-Input thumbnails fail to load | |
Workflow-Builder: Added new navigation options and tooltips | |
Workflow-Builder: Fixed handling of non-create-able model-types | |
Workflow-Builder: Fixed the logic for showing model/version selector in Node details | |
Improved handling of failed inputs that are not marked as failed | |
Onboarding Dialog improvements |