Skip to content

Robovision AI 5.9 release notes

Robovision AI 5.9.1

Release date: May 9, 2025

Bug fixes

  • AI-ADC EfficientNet: The issue where GPUs were not used for training despite being requested has been resolved.
  • EfficientNet, AI-ADC EfficientNet, Multi-Label EfficientNet, SOLOv2: The issue where inference with high-throughput has returned 500 internal server error on /predict request has been resolved.
  • Performance issues in the training center caused by duplicate endpoint requests, particularly in projects with many training processes or metric-heavy algorithms like YOLOv5, have been resolved.

Known limitations

General

  • The Robovision AI UI has been optimized for the browser size of 1920x1200.
  • It is not possible for users to duplicate a tab if the environment is using the HTTP protocol. This does not affect environments using HTTPS.
  • If the storage is full, some functions may become unavailable, for example, saving annotation or training a model.
  • Upgrades are only supported to the next minor version.
  • When upgrading to a new version, new algorithm versions are installed, while preserving the existing ones. Existing projects will be updated to utilize the new algorithms.
  • If the terms and conditions change, you will not be prompted to read and accept them again.

  • Notifications may be missing in specific cases, such as:

    • When copying of annotations fails because the storage is almost full.
    • When an inference setup import fails due to nearly full storage.

  • The Show all datasets button on the project details page redirects to the label center instead of the dedicated Datasets page.

Browser support

  • The Robovision AI platform has been designed and validated for Google Chrome 115 or later.
  • You can transfer data between Robovision AI platforms only in Chromium-based browsers (Google Chrome, Microsoft Edge, and more).
  • In Google Chrome, training metrics and a final validation score might not be displayed if an ad blocker is used. To prevent this, mark the Robovision AI URL as trusted in the Google Chrome settings.
  • In Google Chrome, the thumbnail view can load slowly if you have password managers like Lastpass activated. To prevent this, disable the extension or work in a guest profile window.
  • When using Google Chrome, you might experience issues in the label center. As a quick fix, reset your browser settings.

Backward compatibility

  • Backward compatibility with Robovision AI versions 3.x and 4.x is not foreseen. But upon request, Robovision will provide the needed assistance and conversion scripts.

Branding and brand assets

  • Upon upgrading to a newer version, Robovision AI may include new UI text that has not been customized. To ensure alignment, export the template containing the UI text and review for any necessary updates (see step 3 in Change brand assets).

Project management

  • You can't delete a project if it has running processes.

Data recording

  • Samples recorded while the recording session is stopping are missing from the recording session logs.

Data upload

  • In general and with respect to the model testing functionality, the following image types are supported: JPG, JPEG, PNG, 8-bit TIF and TIFF, BMP, GIF, including a 4th transparency channel. EXIF orientation metadata is supported.
  • File names must contain only Latin (ASCII) characters. There are known rendering, packing, and backup/restore issues with file names/object keys that contain non-Latin characters.
  • You will be logged out of the platform if your data import has been in the Uploading stage for more than an hour.
  • Limited functionality projects: You can still import data if the label center is empty.

Labeling

  • 2D labeler performance (in particular the Magnetic lasso and Grab cut tools) may degrade for high-resolution images.

  • You may face the following issues when using the Prediction tool in the label center:

    • The prediction process gets hidden if you leave the label center. To view it again, click the Prediction tool.
    • The predicted labels are not added to the samples as the prediction process is running.
    • The Status section is not updated as the prediction process is running.

  • If you select more than 100 samples in the label center, the Selected annotations panel on the right includes no information about individual samples. As a result, you will not be able to copy or delete individual annotations.

  • When copying or replacing annotations for multiple samples, the same source should be selected for each sample (for example, annotations from one particular member or one prediction job).
  • EfficientNet: When you re-annotate already annotated samples in bulk, you will not be notified that existing annotations will be overwritten.

Pagination

  • In all views except the single view, a maximum of 100 samples is displayed per page. If you select samples and then go to another page, these samples will no longer be selected.
  • In the test center, you can't select samples beyond the current page (more than 100 samples). The same applies when you view samples in a dataset.

Classes

  • In the label center, it is possible to create a class with a name that is just a space.
  • The class name is not saved if you collapse the Classes section without pressing Enter.

Charts

  • Class distribution:

    • Label and test centers: If you apply a filter when the chart is already open, the filter isn't considered. To refresh the chart, click Clear in the upper-right corner of the Chart setup panel, and then set up the chart again.

  • Confusion matrix:

    • In both the UI and exported CSV files, percentages are rounded to whole numbers.
    • When setting up the confusion matrix, you can select the same user for both "Labels by (x-axis)" and "Labels by (y-axis)".

    • Label center:

      • If you apply a filter when the chart is already open, the filter isn't considered. To refresh the chart, click Clear in the upper-right corner of the Chart setup panel, and then set up the chart again.
      • When you generate the chart for filtered data, some information in the chart might be inaccurate.

    • Test center:

      • YOLOv8+: For some samples, no model annotations are displayed in the chart, even though the annotations exist.
      • The chart doesn't consider the filters you have applied—the number and percentage of samples labeled with a certain class is incorrect. To view the actual number of samples, you'll need to click the needed cell.
      • SOLOv2, YOLOv5: When a test compares two models or different parameters of the same model, more than one matrix may be generated.

  • Wafer map:

    • The wafer map chart is optimized for the browser size of 1920x1080 in full-screen mode. To enter full-screen mode, press F11.
    • When the wafer map contains more than 100,000 samples, selection and filtering within the chart may affect performance and increase the loading time.

Model training

  • PIDNet: When labeling a consistent area with the ignore class, the model might predict every class in those areas. To avoid this, ensure there is some variation in the areas labeled as ignore.
  • PIDNet: If the entered batch size is too large, no validation message is displayed, and the training cannot be started.
  • YOLOv8+: Subtle differences in score calculations during model validation and in the test center may result in discrepancies between the scores, even when using the same settings and data.
  • During training setup, the Add model button is enabled even if there are no models available for transfer learning.
  • When you stop and delete a training that had a custom name, the next training you set up will have the same custom name instead of the default one.

Model testing

  • If you test a model on tagged data without comparison, the test data in the test center may disappear after you unassign that tag. However, the training center will still display the correct number of samples used in the test.
  • Filter by class returns no results if the class exists in the ground truth but is absent in the predictions.
  • YOLOv5: If a test is set up with a confidence threshold of 0, it will automatically run with a threshold of 0.001 instead. This temporary workaround prevents the test from failing.
  • The test process will not start if the test name contains exactly 255 characters.

  • Tests may fail if:

    • The dataset includes samples that have been deleted.
    • You set the confidence threshold to 1 for both models when testing on tagged or imported data. The system requires at least one predicted example to compute metrics.

  • When you export test results, the name of the CSV file contains the project and test names from the time the test was run.

Cameras

  • WebRTC stream may fail to work with Basler emulator cameras.

Inferences

  • You can run only one camera inference at a time.
  • You can't rename a running inference. Wait for the inference to stop or stop it manually before renaming.
  • If you stop an inference within the first minute of this first run, inference logs will not be available.
  • SOLOv2: During inference setup, you can set the inference parameters that are outside of the available range. Despite this, the inference will run with the parameters within the range.
  • It is possible to delete a stopping inference from the project details page.
  • You cannot restart inferences in the Failed status. To proceed, stop the inference and start it again.
  • In an inference setup, repeatedly switching between imported models and then saving the setup may cause the platform to crash.
  • If your license expires, you cannot access the inference center to view details of inferences started while the license was active. Renew the license to regain access.

  • Inference monitoring:

    • Opening a project with limited functionality from the Inference monitoring page results in an error. To open the project, go to the Projects module and search for the project directly.
    • On the Inference monitoring page, the same inference metric type be listed multiple times. This can occur after an upgrade when projects are updated to the new algorithm version, so both the old and new versions of the algorithm remain on the platform.

Import and export of data

  • If the import or export process is interrupted, it will restart from the beginning instead of resuming, potentially resulting in multiple notifications about the process start.

  • Data transfer from Robovision Edge or connected Robovision AI platform:

    • On the target Robovision AI platform, you won't receive notifications about data upload.
    • On the project details page of the target Robovision AI platform, you can't stop the data upload.
    • Exporting large numbers of samples to the connected platform may fail. To avoid this, export no more than 200,000 samples at a time.
    • If the bandwidth is below 8 Mbit, the export to the connected platform may fail.
    • If you delete an import while it's still in progress, some imported samples may not be deleted. These samples will appear in the label center, but they won't be associated with any import.
    • EfficientNet: During samples export, some samples may incorrectly show empty annotations. Once the data transfer is complete, the annotations will be updated, and the issue will no longer appear.
    • When data upload is in progress, the number of samples displayed above the sample preview in the label center may be out of sync with the number shown in the Imports group of the Filter panel.

  • Inference setups:

    • Imported models aren't displayed on the training details page and the Training center section of the project details page because trainings—not models—are shown there. However, you will still be able to select imported models for transfer learning, to set up a test, and more.
    • YOLOv8+ instance segmentation: Import of inference setups with YOLOv11 models may fail, especially if the network is interrupted. In most cases, the issue resolves itself once the network stabilizes.
    • You can import several models with the same name.

  • Samples:

    • If you don't change the default name, all exports in your project will share the same name.
    • Sample export or import, especially for more than 100 samples, may become stuck. To resolve this, manually resume the process.

Robovision AI 5.9

Release date: April 1, 2025

What's new

Smart mask tool

  • You can now perform one-click segmentation with the Smart mask tool. Powered by the SAM (Segment Anything Model), it automatically refines and generates highly precise object boundaries significantly reducing the time required to perform segmentation. See Smart mask tool.

Inference

  • On the Camera management page of the System module, you can now configure a data emulator. A data emulator allows using imported and tagged data for testing inference.
  • The new Inference monitoring page provides a platform-wide overview of all inferences, allowing you to monitor key metrics and analyze their status across your platform. See Inference monitoring.
  • You can now export and import camera inference setups between connected Robovision AI platforms. Note that when importing a camera inference setup to another platform, the camera configuration will not be included.

Camera calibration

  • You can now perform 2D camera calibration for accurate geometric measurements in industrial quality control and robotics. By capturing images of a printed calibration target, the system maps image pixels to real-world geometry for precise 2D scene measurements. See Calibrate cameras.

Threshold optimization

  • Automatically determine the best confidence thresholds for each class based on your model, annotated dataset, and target unknown rate. Predictions below the threshold are assigned the unknown class, improving accuracy in predictive labeling, tests, and inference. See Threshold optimization.

Data emulators

On the Camera management page of the System module, you can now configure a data emulator. A data emulator allows using imported or tagged data for testing inference. Camera emulators are unavailable for testing inference in custom algorithms for version 5.8 and earlier.

Other improvements

  • Password managers can now auto-fill password fields.
  • Predictive labeling setups can be deleted.

Bug fixes

Project management

  • Previously, deleting a project with models or datasets that were used in other projects resulted in an incomplete deletion. Now, you can't delete a project that contains shared resources.

Data recording

  • The recording session progress now updates consistently both on the project details page and in the recording center.

Labeling

  • Deleting selected predictions on a sample using the delete button in the right panel now removes only the selected predictions, not all predictions for the sample.
  • In YOLOv8+ instance segmentation, the masks2segments function now preserves all contours of a disjoint instance mask instead of selecting only the largest one, thus preventing data loss.
  • The count of filtered results now correctly matches the number of filtered samples when switching views in the label center.
  • Previously, after model testing and deleting annotations from samples, filtering by unannotated samples and running a prediction would fail to add annotations to those samples. Now, the Prediction tool correctly adds annotations in this scenario.

  • Wafer map:

    • Class labels on the wafer map now match the ones in the thumbnails view.
    • The wafer map now correctly reflects applied filters.

Classes

  • Previously, system classes from imported annotated data were missing from the Classes page until the project's label center was opened. This issue is now resolved.

Datasets

  • When you rename datasets on the project details page, their names now must be unique.
  • If you can't create a dataset due to missing annotations from a specific member, the tooltip now clearly explains why.

Model training

  • Previously, training could fail if images in the validation set were larger than the maximum dimensions computed from the training set. This issue is now resolved.

Model testing

  • AI-ADC EfficientNet: Model tests now ignore the unknown class, and it isn't considered in the test result.
  • After annotated data is imported in the test center, its annotations and classes no longer appear in the label center.

Inference

  • You can no longer use the same camera in multiple inferences simultaneously.
  • When you import an inference setup from the connected platform, the Import list now shows inference names instead of inference IDs.

Import and export of data

  • Previously, exports to the connected platform could fail due to low bandwidth, displaying an unclear error message. Now, the error message clearly states that the failure is due to a slow or interrupted connection, and a retry mechanism has been implemented.
  • Inference setups now use definition IDs instead of version numbers to determine compatibility. This change allows seamless exchange across minor versions when the definition ID remains unchanged.

Other bug fixes

  • Data imports now remain visible on the project details page, even if you navigate away during the upload process.
  • Notifications now appear correctly if the connection between two Robovision AI platforms is lost during a data transfer.
  • Previously, using the & character in your password could trigger the following error: An error occurred while accepting Terms and Conditions. This issue has been resolved.
  • User avatar colors now update correctly in the dark mode without requiring a refresh.

Known limitations

General

  • The Robovision AI UI has been optimized for the browser size of 1920x1200.
  • It is not possible for users to duplicate a tab if the environment is using the HTTP protocol. This does not affect environments using HTTPS.
  • If the storage is full, some functions may become unavailable, for example, saving annotation or training a model.
  • Upgrades are only supported to the next minor version.
  • When upgrading to a new version, new algorithm versions are installed, while preserving the existing ones. Existing projects will be updated to utilize the new algorithms.
  • If the terms and conditions change, you will not be prompted to read and accept them again.

  • Notifications may be missing in specific cases, such as:

    • When copying of annotations fails because the storage is almost full.
    • When an inference setup import fails due to nearly full storage.

  • The Show all datasets button on the project details page redirects to the label center instead of the dedicated Datasets page.

Browser support

  • The Robovision AI platform has been designed and validated for Google Chrome 85 or later.
  • You can transfer data between Robovision AI platforms only in Chromium-based browsers (Google Chrome, Microsoft Edge, and more).
  • In Google Chrome, training metrics and a final validation score might not be displayed if an ad blocker is used. To prevent this, mark the Robovision AI URL as trusted in the Google Chrome settings.
  • In Google Chrome, the thumbnail view can load slowly if you have password managers like Lastpass activated. To prevent this, disable the extension or work in a guest profile window.
  • When using Google Chrome, you might experience issues in the label center. As a quick fix, reset your browser settings.

Backward compatibility

  • Backward compatibility with Robovision AI versions 3.x and 4.x is not foreseen. But upon request, Robovision will provide the needed assistance and conversion scripts.

Branding and brand assets

  • Upon upgrading to a newer version, Robovision AI may include new UI text that has not been customized. To ensure alignment, export the template containing the UI text and review for any necessary updates (see step 3 in Change brand assets).

Project management

  • You can't delete a project if it has running processes.

Data recording

  • Samples recorded while the recording session is stopping are missing from the recording session logs.

Data upload

  • In general and with respect to the model testing functionality, the following image types are supported: JPG, JPEG, PNG, 8-bit TIF and TIFF, BMP, GIF, including a 4th transparency channel. EXIF orientation metadata is supported.
  • File names must contain only Latin (ASCII) characters. There are known rendering, packing, and backup/restore issues with file names/object keys that contain non-Latin characters.
  • You will be logged out of the platform if your data import has been in the Uploading stage for more than an hour.
  • Limited functionality projects: You can still import data if the label center is empty.

Labeling

  • 2D labeler performance (in particular the Magnetic lasso and Grab cut tools) may degrade for high-resolution images.

  • You may face the following issues when using the Prediction tool in the label center:

    • The prediction process gets hidden if you leave the label center. To view it again, click the Prediction tool.
    • The predicted labels are not added to the samples as the prediction process is running.
    • The Status section is not updated as the prediction process is running.

  • If you select more than 100 samples in the label center, the Selected annotations panel on the right includes no information about individual samples. As a result, you will not be able to copy or delete individual annotations.

  • When copying or replacing annotations for multiple samples, the same source should be selected for each sample (for example, annotations from one particular member or one prediction job).
  • EfficientNet: When you re-annotate already annotated samples in bulk, you will not be notified that existing annotations will be overwritten.

Pagination

  • In all views except the single view, a maximum of 100 samples is displayed per page. If you select samples and then go to another page, these samples will no longer be selected.
  • In the test center, you can't select samples beyond the current page (more than 100 samples). The same applies when you view samples in a dataset.

Classes

  • In the label center, it is possible to create a class with a name that is just a space.
  • The class name is not saved if you collapse the Classes section without pressing Enter.

Charts

  • Class distribution:

    • Label and test centers: If you apply a filter when the chart is already open, the filter isn't considered. To refresh the chart, click Clear in the upper-right corner of the Chart setup panel, and then set up the chart again.

  • Confusion matrix:

    • In both the UI and exported CSV files, percentages are rounded to whole numbers.
    • When setting up the confusion matrix, you can select the same user for both "Labels by (x-axis)" and "Labels by (y-axis)".

    • Label center:

      • If you apply a filter when the chart is already open, the filter isn't considered. To refresh the chart, click Clear in the upper-right corner of the Chart setup panel, and then set up the chart again.
      • When you generate the chart for filtered data, some information in the chart might be inaccurate.

    • Test center:

      • YOLOv8+: For some samples, no model annotations are displayed in the chart, even though the annotations exist.
      • The chart doesn't consider the filters you have applied—the number and percentage of samples labeled with a certain class is incorrect. To view the actual number of samples, you'll need to click the needed cell.
      • SOLOv2, YOLOv5: When a test compares two models or different parameters of the same model, more than one matrix may be generated.

  • Wafer map:

    • The wafer map chart is optimized for the browser size of 1920x1080 in full-screen mode. To enter full-screen mode, press F11.
    • When the wafer map contains more than 100,000 samples, selection and filtering within the chart may affect performance and increase the loading time.

Model training

  • PIDNet: When labeling a consistent area with the ignore class, the model might predict every class in those areas. To avoid this, ensure there is some variation in the areas labeled as ignore.
  • PIDNet: If the entered batch size is too large, no validation message is displayed, and the training cannot be started.
  • YOLOv8+: Subtle differences in score calculations during model validation and in the test center may result in discrepancies between the scores, even when using the same settings and data.
  • During training setup, the Add model button is enabled even if there are no models available for transfer learning.
  • When you stop and delete a training that had a custom name, the next training you set up will have the same custom name instead of the default one.

Model testing

  • If you test a model on tagged data without comparison, the test data in the test center may disappear after you unassign that tag. However, the training center will still display the correct number of samples used in the test.
  • Filter by class returns no results if the class exists in the ground truth but is absent in the predictions.
  • YOLOv5: If a test is set up with a confidence threshold of 0, it will automatically run with a threshold of 0.001 instead. This temporary workaround prevents the test from failing.
  • The test process will not start if the test name contains exactly 255 characters.

  • Tests may fail if:

    • The dataset includes samples that have been deleted.
    • You set the confidence threshold to 1 for both models when testing on tagged or imported data. The system requires at least one predicted example to compute metrics.

  • When you export test results, the name of the CSV file contains the project and test names from the time the test was run.

Cameras

  • WebRTC stream may fail to work with Basler emulator cameras.

Inferences

  • You can run only one camera inference at a time.
  • You can't rename a running inference. Wait for the inference to stop or stop it manually before renaming.
  • If you stop an inference within the first minute of this first run, inference logs will not be available.
  • SOLOv2: During inference setup, you can set the inference parameters that are outside of the available range. Despite this, the inference will run with the parameters within the range.
  • It is possible to delete a stopping inference from the project details page.
  • You cannot restart inferences in the Failed status. To proceed, stop the inference and start it again.
  • In an inference setup, repeatedly switching between imported models and then saving the setup may cause the platform to crash.
  • If your license expires, you cannot access the inference center to view details of inferences started while the license was active. Renew the license to regain access.

  • Inference monitoring:

    • Opening a project with limited functionality from the Inference monitoring page results in an error. To open the project, go to the Projects module and search for the project directly.
    • On the Inference monitoring page, the same inference metric type be listed multiple times. This can occur after an upgrade when projects are updated to the new algorithm version, so both the old and new versions of the algorithm remain on the platform.

Import and export of data

  • If the import or export process is interrupted, it will restart from the beginning instead of resuming, potentially resulting in multiple notifications about the process start.

  • Data transfer from Robovision Edge or connected Robovision AI platform:

    • On the target Robovision AI platform, you won't receive notifications about data upload.
    • On the project details page of the target Robovision AI platform, you can't stop the data upload.
    • Exporting large numbers of samples to the connected platform may fail. To avoid this, export no more than 200,000 samples at a time.
    • If the bandwidth is below 8 Mbit, the export to the connected platform may fail.
    • If you delete an import while it's still in progress, some imported samples may not be deleted. These samples will appear in the label center, but they won't be associated with any import.
    • EfficientNet: During samples export, some samples may incorrectly show empty annotations. Once the data transfer is complete, the annotations will be updated, and the issue will no longer appear.
    • When data upload is in progress, the number of samples displayed above the sample preview in the label center may be out of sync with the number shown in the Imports group of the Filter panel.

  • Inference setups:

    • Imported models aren't displayed on the training details page and the Training center section of the project details page because trainings—not models—are shown there. However, you will still be able to select imported models for transfer learning, to set up a test, and more.
    • YOLOv8+ instance segmentation: Import of inference setups with YOLOv11 models may fail, especially if the network is interrupted. In most cases, the issue resolves itself once the network stabilizes.
    • You can import several models with the same name.

  • Samples:

    • If you don't change the default name, all exports in your project will share the same name.
    • Sample export or import, especially for more than 100 samples, may become stuck. To resolve this, manually resume the process.