Release Notes

6.1.3 - January 22, 2025¶

New Features¶

Added new environment variable to help with debugging videos that fail to import. Setting REDACTOR_LOADVIDEO_ERROR_DUMPS=1 (increase 1 to how many debug folders/videos you want to keep on system) will copy the original and the converted videos (if any) into a local folder, mark this folder with the session ID, drop additional information (request etc) into the folder, and delete older folders if we run out of space/capacity.

Updates¶

Preventing the contact-support dialog from being submitted multiple times in a row, present a status message while transmitting.
Remove unknown tracks from media during import.

Bug Fixes¶

(Desktop) neither the Add Media button nor the File->Open menu would load media. Only dragging and dropping worked. Caused by Electron update in 6.1.2

6.1.2 - January 14, 2025¶

New Features¶

Upgrade Electron to latest version.
Pressing F1 on keyboard when in editor will show a list of keyboard shortcuts.

Updates¶

(API) Add support for submit MD5 of content if needed by S3

Bug Fixes¶

(Server) Fix issue with new installations getting stuck on initial setup screen.
(Desktop) Stepping through a video frame-by-frame could show the boxes in a different location than the rendered output. They were off by one frame.

6.1.1 - December 19, 20024¶

New Features¶

Replace native browser dialogs with application specific ones.
Allow uploading of media into current folder. Previously, all uploads would be placed in the root location.
Updated context menu on main page to work better with bulk operations.

6.1.0 - December 13, 2024¶

New Features¶

Folder Support - Media can now be moved and grouped into folders on the main page.

Updates¶

Improved handling of interlaced videos and those with odd resolutions during import conversion stage.

Bug Fixes¶

Improved support of filenames with uncommon characters (like #)
Fixed internal issue with persisted storage and write queueing causing issues (introduced in v6.0.5)

6.0.7 - December 10, 2024¶

New Features¶

(API) add ability to ignore SSL/TLS certificate issues for webhook URIs. Must be specified in request body as metadata etc_event_webhook_ignore_certificate_errors: true

Updates¶

(API) Add skipOnError request option für STATE_DROP for debugging purposes (error state is kept for querying)

Bug Fixes¶

(Desktop) The preview button was not working properly and would only show a black area instead of the exported media. This depends on the locally exported file still being available at viewing time.

6.0.6 - December 2, 2024¶

Updates¶

Improved error handling for speech transcription processing (inconclusive before).

Bug Fixes¶

(API) STATE_DROP was being skipped if sessionId wasn't provided in request body.

6.0.5 - November 11, 2025¶

New Features¶

(API) Ability to custom name the exported file when written back to storage/bucket using videoContext.outputContext.exportKey = "my exported file.mp4" in the request.
New keyboard shortcut [W] will cycle through the redaction objects on the screen
New keyboard shortcut [Delete] will delete the currently selected object
New environment variable REDACTOR_GPU_IGNORE. Defaults to 0. Set to 1 to disable the usage of any GPU in case of incompatibilities.

Bug Fixes¶

Fixed issue with audio redaction regions not muting during playback in editor.
Update the label used for Chinese language in speech transcription.

6.0.4 - October 28, 2024¶

New Features¶

New keyboard shortcuts added for manual editing. [A] and [D] keys to move backward and forward one frame, the same way as the left/right arrow keys do. After modifying the position or size of an existing redaction box, pressing the [S] key will apply the change to a single frame, like click the button of the same name.

Updates¶

If GPU is present, skip building the models that are not used in the UI.
Show the Export Range option in the Render and Export modal, even if no redactions have been specified. This is useful to just trim the video.

Bug Fixes¶

Displayed video editing features for audio files.

6.0.3 - October 1s7, 2024¶

Updates¶

AWS/S3 can now be used without explicitly providing credentials via Redactor environment variables. Counting on properly set up system for AWS (environment etc).

6.0.2 - October 15, 2024¶

Updates¶

Enhanced crash handling for processing videos.

Bug Fixes¶

Fixed issue with audio truncation.

6.0.1 - October 10, 2024¶

New Features¶

Audio only redaction is now possible.
(API) Audio regions can be passed in via API request

Updates¶

(API) Provided session identifier is now emitted in API progress messages

6.0.0 October 7, 2024¶

New Features¶

Head Detection added as option to auto-detect - replaces face detection, which is still available via the API.
Display custom metadata from API in the video list.
Search box added to projects page. It will search filenames and custom metadata if provided via API calls.
(API) support for specifying export video range in seconds.
(API) Emitted render event now carries video range as timestamps in seconds.

Updates¶

Rename Videos route to Media in Admin section
Removed Faces from auto-detection choices and replaced with heads.
Attempt to pull new license from remote licensing server if local one is expired, instead of becoming unlicensed.
Head/face box expansion can be specified via environment variable REDACTOR_PROCESS_(HEAD|FACE)BOX_EXPAND, API and also profiles.
Export dialog now displays only options which mater, e.g. audio redaction options will not be displayed if the media doesn’t contain audio.

Bug Fixes¶

(API) Filename/URI sanitation in the export URI needed (%, etc.)
Display the export range for audio-only redactions.
Fixed bug with exported video (missing last frame and playback issues).
Fix socket transport disconnection trigger on destroy (closing the wrong session).

5.2.13 - August 8, 2024¶

Updates¶

Custom metadata presentation support added.
(API) Legacy processConfig options supported again to mitigate error handling/crashes caused by this.
(API) Custom metadata resolution on intake.
Show "picture" icon over thumbnail on Projects page when media is an image.
Video list action buttons are disabled when video is in Use.
Contact Support modal wording and layout changed.

Bug Fixes¶

(API) Use filename from Content-Disposition header if available by the server.
Modifying static boxes now only shows Apply and Cancel buttons.
v5.2.12 introduced an issue with displaying the list of videos in the Admin Videos section.
(API) v5.2.12 introduced an issue with applying ACL

5.2.12 - July 16, 2024¶

New Features¶

(API) Ad-hoc, static redaction regions can now be defined in the API request body as part of the renderConfig in redactionTransientData. These regions will be overlaid with any other existing redactions, but they will not be persisted like other redaction regions. This is useful for redacting known areas of a video like timestamps or logos. The transient redaction coordinates can be specified as either absolute pixels or relative percentages of the video's dimensions [0...1]. See the example below for more details.

{
    "inputUri": "https://127.0.0.1:60750/media/small.mp4",
    "features": [
        "FACE_DETECTION",
        "LICENSE_PLATE_DETECTION",
        "MEDIA_RENDER",
        "STATE_DROP"
    ],
    "videoContext": {
        "renderConfig": {
            "redactionIntensity": "MEDIUM",
            "redactionShape": "ELLIPSE",
            "redactionStyle": "BLUR",
            "videoPreset": "FAST",
            "videoQuality": 0.55,
            "redactionTransientData": {
                "static": [
                    {
                        "x": 8,
                        "y": 4,
                        "width": 32,
                        "height": 16
                    },
                    {
                        "x": 0.9,
                        "y": 0.95,
                        "width": 0.098,
                        "height": 0.075,
                        "relative": true
                    }
                ]
            }
        }
    },
    "outputUri": "file:///C:/example/my-bucket/redacted.mp4",
}

(Embedded UI) Initial work was completed to use an iframe to display the editor from an html page hosted on the Redactor server. More details, documentation, and examples are coming soon, but see the following for a very basic example HTML page that embeds Redactor via iframe. The integration supports sending and receiving events to/from the iframe through postMessage and the redactor.invoke.* events.

<!DOCTYPE html>
<html>
    <head>
        <title>{REDACTOR_EMBEDDED_TITLE}</title>
    </head>
    <body>

        <!-- __DEMO_BEGIN__ -->
        <div>
            <button id="closeBtn" onClick="onClose()" style="color: darkblue">Close</button>
            <button id="exportBtn" onClick="onExport()" style="color: darkgreen" >Export</button>
        </div>
        <!-- __DEMO_END__ -->

        <iframe
            src="{REDACTOR_EMBEDDED_IFRAME_SRC}"
            title="{REDACTOR_EMBEDDED_IFRAME_TITLE}"
            style="width: 100vw; height: 100vh"
            id="iframe_redactor_embedded">
        </iframe>

        <script>
            console.log("REDLOG-I: subscribing to iframe messages ...");
            window.addEventListener("message", e => {
                console.log("REDLOG-I: got iframe message", JSON.stringify(e.data));
            });
            // __DEMO_BEGIN__

            const iframe = () => document.getElementById("iframe_redactor_embedded");

            function onClose() {
                iframe().contentWindow.postMessage({
                    type: "redactor.invoke",
                    invoke: "sendCommand",
                    command: "close",
                }, "*");
            }

            function onExport() {
                iframe().contentWindow.postMessage({
                    type: "redactor.invoke",
                    invoke: "sendCommand",
                    command: "export",
                }, "*");
            }

            // __DEMO_END__
        </script>
    </body>
</html>

Updates¶

(Server) The download filename no longer contains the session id as a prefix.
Show confirmation dialog when a user attempts to remove a video on the Projects page.

Bug Fixes¶

(Server) The filename for the redacted download was incorrectly missing the original filename.

5.2.11 - July 11, 2024¶

New Features¶

(API) Metadata can now be provided in prepareConfig. Use etc_media_title to set the display name of the media in the Projects page and for export name purposes. If not provided, the display name will be the filename.
```
"prepareConfig": {
    "metadata": {
        "etc_media_title": "my custom title",
    }
}
```
(API) Allow a custom “Save” button to be displayed in the normal Redactor Server UI. This feature works alongside the new browser event webhook mentioned below and will emit events when a user clicks the Save button. This could be used by your team to signal the user's intent to push the redacted video to your backend service.

(API) Redactor server can send certain browser events to a webhook which is specified via ENV or in an API request body during MEDIA_PREPARE. Full documentation and an example will be added to this site soon. To specify the webhook via ENV, set a System environment variable REDACTOR_EVENT_WEBHOOK (E.g. REDACTOR_EVENT_WEBHOOK=http://127.0.0.1:7737/). Or, to specify the webhook via API request body, use the etc_event_webhook item in the prepareConfig.metadata object:

"videoContext": {
    "prepareConfig": {
        "metadata": {
            "etc_event_webhook": "http://127.0.0.1:7737/",
            "etc_customize": {
                "export": {
                    "download": "hide",
                    "save": "close"
                },
                "session": {
                    "remove": "hide",
                    "deleteRedactionData": "show"
                }
            },
            "etc_media_title": "my custom title"
        }
    }
},

Updates¶

Display an error message if the in-app "Contact Support" function does not submit properly. This is typically only an issue for those with a Redactor server behind a restrictive firewall.
Show confirmation dialog when a user attempts to delete redaction data from a video on the Projects page.
(Embedded UI) Saving image from current video frame now works when using Shadow DOM

Bug Fixes¶

Timestamps and tick marks in audio waveform were not always visible.
The path used for the last export was not being remembered in the desktop version.

5.2.9 - June 10, 2024¶

New Features¶

(API) Synchronize a redaction session's data with a different bucket/location. If a session:// is given for the input URI and a bucket for the output URI, the entire state of the session will be saved to the outputUri. This is useful if you want to create a backup of the session data or move it off the server/S3/Azure before shutting it down. You will be able to restore the session from the data stored in the outputUri.
Safari browser is now supported. We haven’t used that browser in the past, so if you find rare glitches on your end please let us now. Tested against the latest Safari 17.4.1 (MacOS 14.4.1 on M1).

Updates¶

(API) Improved bucket synchronization efficiency: The manifest.json left in buckets now contains checksums will avoid media files being retransmitted if they already exist on the server.
(API) forceConversion also enabled by default in the API. It is still possible to disable conversion at MEDIA_PREPARE via forceConversion: false if needed, but we recommend to convert always, since it fixes a lot of glitches we’ve seen in the past due to irregular video data, in a broad sense.
(API) Context menu enabled and can be customized by the Redactor embedded API constructor call.
(Embedded UI) Console output can now be filtered by a log level and the messages prefixed for distinct handling, configurable in the SighthoundRedactorEditor constructor options.
(Embedded UI) Made the SighthoundRedactorEditor.cleanup() async so it’s safe to continue when the editor is truly closed; this was possible before for initiating it and then wait for the exit event, yet the promise is an integrated and a more convenient choice
Websockets connectivity improved. In case of connection loss, resumption should be transparent and non-interruptive for an active user session, up to a certain timeout or a server restart. Pending messages will be replayed.
Windows Installer didn’t store custom port numbers on fresh installs. This only happened when no former data directory existed. Now it gets created first thing.

Bug Fixes¶

(Embedded UI) Download and preview links in the video list were broken. In case downloadOnly is used for redirecting downloads the associated event now carries the URL which would have been used directly.
(Embedded UI) The renderConfig event wasn’t truly reflecting all of the choices possible in the export dialog.
(API) Audio waveform package's global instance prevented the script from being reloaded.
(API) maskAudio regions provided in the renderConfig was removing custom audio regions. This has been adjusted and the regions are now rendered correctly.
Audio mute button didn’t restore the formerly set volume.
Scrollbars were always showing in audio transcripts for Chrome & Edge browsers.
Image loading didn’t clear the progress dialog.
Export save path wasn’t remembered in the desktop version.

5.2.1 - April 18, 2024¶

New Features¶

(Embedded UI) “Save” button option for embedded UI.
(Embedded UI) Download Event - redactor.export.download event is emitted when users click either of the Download buttons and it includes the sessionId for the selected video.
Set REDACTOR_DIAG_PROBE_MODE default to “always”.
(API) /api/v11/videos:list endpoint to get an array of videos loaded into Redactor and their “rendered” status. It can be filtered based on sessionId or projectId.
(Embedded UI) Add configurable “Save” button to Projects page items
Allow specifying different outputUri for API calls using session:// as inputUri. The outputUri can be different than those used (if provided) in the initial MEDIA_PREPARE call.
New "/api/v11/videos:list" endpoint to query a list of videos loaded into Redactor. It can be filtered based on sessionId or projectId.
FFprobe is run for all imported videos and results are logged.

Updates¶

Improve the (optional) ignoring of SSL certificate errors for video imports from remote resources.
(Embedded UI) Scramble intensity events are now emitted in redactor.export.submit event in “maskAudioParams”. Intensity value will be either LOW, MEDIUM, or HIGH.

Bug Fixes¶

(API) The timeout for internal API operations has been removed. Previously, the API timeout would be the same as the one set to “Disconnect idle users”.
(Embedded UI) Video Preview was not appearing on top of buttons.
(Embedded UI) Fix dropdown not working in Export Dialog (Shadow DOM).
Filenames with emoji and some Unicode characters would crash the render/export.
Fixed potential issues when canceling an import operation.
CSS updates for button transitions.
Audio waveform tick marks were not showing up in light mode.
Various Node and Package updates.
Person checkbox could not be selected/deselected in Objects List.
Fix export for videos that contain audio tracks with blocks of missing packets.

5.2.0 - March 7, 2024¶

New Features¶

Modify the size or location of an object for only one frame with the "Apply to Single Frame" button. The previous behavior of retracking the object based on the changes is still available and is now called "Apply and Track".

Updates¶

Start and stop times can now be specified as hh:mm:ss when exporting a portion of the video.
Imported videos will now be re-encoded by default. This should prevent any synchronization issues between the video and rendered boxes during editing. Holding the SHIFT key down before importing a new video will bypass conversion.

Bug Fixes¶

Fixed issue with Export Range not working properly
Fixed a bug with rendering not working correctly on multi-export
Progress reporting was not working properly for video conversion/encoding at import time
(API) Fixed issue with API service not starting after new license activation
(API Embedded UI) React context menu click not working in Shadow DOM

5.1.0 - January 13, 2024¶

Updates¶

Improve detections for fast-moving objects with Body/Dash Cam profile
(API) AZURE_STORAGE_ACCOUNT and AZURE_STORAGE_ACCESS_KEY are not required to be specified if using SAS for input/output URIs
(API) AZURE_STORAGE_ACCOUNT value does not need to match the output Azure Storage Account when using SAS
(API Embedded UI) Shadow DOM beta
- KNOWN ISSUE: Can’t select options after right-clicking on detections

Bug Fixes¶

Fix white screen issue when user with Supervisor group tries to access the Admin section
(API) Images can now be exported

5.0.6 - December 22, 2023¶

New Features¶

Added “Invert Regions” to Audio Redaction to "unmute" a portion of previously selected audio

Updates¶

Update the render options in the bulk export dialog to include audio redaction strength & "default" settings
(API) API export could lock a video as "in-use" if consecutive export calls were made to the same outputUri

Bug Fixes¶

Previewing already exported video did not show updated redaction types
Add Media & Convert now properly resets after the file uploads
Fixed “Merge Regions” to only appear if audio redactions can be merged

5.0.5 - December 11, 2023¶

Updates¶

Redaction boxes may disappear on last frame for certain videos
Fixed force conversion not working on drag and drop

API¶

Full export options exposure to the API

5.0.4 - November 27, 2023¶

New Features¶

Audio transcript text can be downloaded from within the Speech panel

Updates¶

Capture additional logs if a video fails to load.
"Add Media" button now shows "Add Media & Convert" when holding SHIFT on keyboard.
Removed country flags from multi-language UI selector in settings.
Removed the detection types from the export settings. What is selected in the view through [Objects] is what will be rendered.
Added missing audio setting into Multi-select export settings.
The multi-select settings overwrite all of the individual media’s export settings, except the ones not presented there, because of the compound nature (detection types, region, …).
Included full timestamp in exported filename instead of a random number.
Do not show export options that won’t apply to the selection: no audio, only images, etc.
Save the multi-export options so it remembers them for next time.
Fix the issue where if multiple videos are selected and one is deleted, the status (5 + selected ) doesn’t change and stays sticky until a video is opened again to clear the state.

Bug Fixes¶

Image redaction export crashes if bounding box touched edge

API¶

(Embedded UI) "active()" method added to embedded UI to notate editor is open
New licensing endpoint: GET http://localhost:${SERVER_PORT}/licensing
Fix issue where invalid projectId format would crash service
Events are emitted with sessionId when editor is entered and exited

4.5.5 - November 9, 2023¶

Bug Fixes¶

Loading or exporting videos would occasionally show errors on certain hardware configurations

Updates¶

Expanded logging to assist with support

5.0.3 - October 24, 2023¶

New Features¶

Show "Unsupported Browser" message for Safari browser
Checkmark now appears beside audio transcript languages that have already been processed

Updates¶

Added "View options" and "Redaction type" filter to Objects Panel
(API Embedded UI) Emit commands and events from the editor

Bug Fixes¶

"Automatically Track" timing issues for boxes

5.0.1 - October 9, 2023¶

Bug Fixes¶

Removed 5-minute timeout for large file uploads
Automatically Track was off for forward and backward passes
Cancelling processing tasks would cause a crash
Fall back to CPU processing mode for GPUs with compute capability < 5

5.0.0 - October 5, 2023¶

New Features¶

Speech detection and transcription for videos
Multi-language support in the UI
Upgrade to Sighthound Gen6 Computer Vision models
Enable CUDA video decoding
New Settings/Help Section
Frame-by-frame editing mode (Hold down SHIFT key before clicking Apply in Edit mode)
(API) "Projects" feature added to the embedded UI

Updates¶

Performance improvements for processing and exporting with GPU
Move Export Options from sidebar into a modal that’s displayed on export
(Docker) Set the timezone in the container to GMT

Bug Fixes¶

Disable video controls when in editing mode to show that playback is prevented
Bug fix for exporting trimmed and scrambled videos
A few log entries were incorrectly displayed as errors

4.5.3 - June 9, 2023¶

Bug Fixes¶

Fixed incorrect build location for GPU model cache if a custom Redactor data folder is configured. This caused the GPU build process to run twice.
Changed the wording in Desktop version to say "Open Exported Image" when an image is redacted instead of a video.
Fixed truncated text in "Sort By" when using Firefox.

Updates¶

Updated Sighthound License Agreement.

4.5.2 - April 14, 2023¶

Bug Fixes¶

Progress bar no longer jumps around when a file is uploading at the same time a video is processing.
Added missing translations for API embedded UI users.

4.5.1 - March 22, 2023¶

New Features¶

Redactor API users can now specify a project-id and location-id for the Operation name. This is helpful when running multiple Redactor API servers behind a proxy or load balancer so you'll know which one served the request. REDACTOR_API_PROJECT_DEFAULT and REDACTOR_API_LOCATION_DEFAULT are the new environment variables. The default value is "0". (Redactor Server only)

Bug Fixes¶

Some NVIDIA GPU users would not see the "Preparing Models" progress dialog after running an auto-detection on a new Sighthound Redactor installation. The "Processing" progress dialog would be displayed instead and would appear not to move for several minutes until the GPU model build finished.
The initial "Preparing Models" stage was taking about 7-10 minutes longer than usual for NVIDIA users due to building unneeded GPU models.

4.5.0 - March 20, 2023¶

New Features¶

Images can now be redacted in the editor.
A new "Recovery Mode" option has been added to allow server administrators to access Redactor in case of a lockout. (Redactor Server only)
Redactor API users can now save/export detection metadata in JSON format. (Redactor Server only)

New Environment variables¶

REDACTOR_PROCESS_FACEBOX_EXPAND - Specifies the scaling factor to use for the face detection boxes generated during auto-detection. Value can be from 1 - 5. Default is 1.5 (50% larger).
REDACTOR_COOKIE_SESSION_NAME - Allows the server administrator to change the default session cookie name from connect.sid to a different value in case of conflict with other applications running on the server. (Redactor Server only)

Updates¶

Several buttons and dialogs in the application have been renamed to include "media" or "images" instead of solely "video".
Password confirmation has been added to the Admin section when creating user accounts or changing passwords.
Password confirmation has been added to the user-facing "Forgot Password" page.
The Admin "Settings" page has been renamed to "Processing" in preparation for future configuration options.
Media imported via API now display their filenames on the Projects page instead of ids.
General code/package updates.

Logging Updates¶

Created a new logs directory to hold all of the log files. This folder can be zipped and sent along with support tickets.
A new sysinfo.json file will be created in the logs directory to provide details on the user's computer specs (CPU, RAM, etc.).
Redactor's Windows Service log file is now written into the logs directory.

Bug Fixes¶

Fixed an issue related to an audio processing crash that would cause the UI to go blank.
The processing icon no longer shows 0 after a successful bulk operation.

Known Issues¶

On the "Forgot Password" reset page, the form will not submit if the "password" and "confirm password" fields don't match (as expected), but no error will be visible on the screen (unexpected). If this issue occurs, retype both passwords and try submitting again.

4.4.5 - January 30, 2023¶

New Features¶

A 4x playback speed option was added to the video player.
The Redactor API has a new ACL feature to specify which Redactor user(s) should be given access to the imported video

Updates¶

The video import/conversion process has been improved to support a wider variety of sources, particularly those with additional non-audio/non-video streams (e.g binary or subtitle tracks).
We squashed a few bugs and made several small updates to keep things running smoothly.

4.4.4 - November 17, 2022¶

New Features¶

Items in the Objects List can be bulk selected and deleted. Hold Shift or Ctrl to select, and right-click to open the menu to delete.

Updates¶

The "beep" tone used for audio redaction in exported video is slightly quieter.
The "Contact Support" form requires an email address to submit.
Username and password fields are now required fields (i.e. can't be blank) throughout the application.

Bug Fixes¶

Removing the SMTP Settings in the Admin page would crash the UI.
The token/link to reset a password was expiring quicker than expected.
Certain errors during login would cause layout issues.

4.4.3 - October 6, 2022¶

Updates¶

For users with NVIDIA GPUs, the wording changed to better reflect what's happening on initial launch when the GPU models are building.
The permissions granted to an Embedded API session/user have been further reduced.

Bug Fixes¶

The "In Progress" message was barely visible when using Dark Mode.

4.4.2 - September 21, 2022¶

New Features¶

VOB containers are now supported.
Admin and Logout buttons were added to the Redactor Server UI.

Bug Fixes¶

Videos with non-square pixels would not render redactions properly. This mainly impacted video from older surveillance systems.
The progress bar displayed after importing a video wasn't updating during the video conversion process.
The hostname, protocol, and port used in the "reset password" email could be incorrect when running Redactor behind a reverse proxy. It's now configurable in the Admin page.

4.4.1 - September 15, 2022¶

New Features¶

A new Dark Mode option is available in the UI.

Updates¶

The "Forgot Password" page received some styling changes.
Specifying a new redaction data storage location which automatically moves the files has been improved.
The core SMTP packages have been updated and work better with secure connections.

Bug Fixes¶

Canceling uploads could potentially crash the application.
Very large videos (15GB+) would occasionally crash on import.
During export, very specific redaction box coordinates could cause a crash during export.
Videos that failed the conversion stage were not being removed from the system.

4.3.1 - August 2, 2022¶

New Features¶

Azure SAS is now supported as input and output URIs for the API.
Azure Gov domains are now supported for API input/output storage.

Updates¶

Video playback performance is greatly improved for videos with lots of detected objects.
The default face detection box size changed to 1.5 instead of 2.

4.2.3 - May 23, 2022¶

Bug Fixes¶

A crash during processing could potentially get stuck in the queue and prevent other redactions from starting.
Redactor API notifications (e.g. completion events) were not always sent if a crash during processing occurred.

4.2.2 - May 2, 2022¶

New Features¶

Users can now select a "Detection Profile" to use when running auto-detection. These will tune the computer vision to best suit those types of videos.
The ability to sort the videos on the main page was added.
A "strength" option was added for the "scramble" audio redaction type.
Redactor API webhooks were not firing for START and COMPLETION event types.

Bug Fixes¶

Unsupported NVIDIA GPUs would crash the service on launch. We now fallback to CPU mode for these cases.

4.2.0 - April 26, 2022¶

New Features¶

Users can now select a "Detection Profile" to use when running auto-detection. These will tune the computer vision to best suit those types of videos.
The ability to sort the videos on the main page was added.
A "strength" option was added for the "scramble" audio redaction type.

4.0.46 - April 26, 2022¶

Updates¶

Exported videos are now smaller due to a change in compression settings.

4.0.45 - April 11, 2022¶

Bug Fixes¶

Exporting videos that had out-of-bounds redaction coordinates would crash. We now scan/correct the coordinates before export.

4.1.0 - March 22, 2022¶

New Features¶

The UI for the main screen has been redesigned.
A new bulk mode was added to allow multiple videos to be imported, redacted, and exported from the main page.
A live tracking feature was added to the Redactor editor to allow a user to follow an object with their mouse pointer during playback to notate where the redaction box should be placed. This is ideal for fast moving objects or those that don't work well with Redactor's built in manual tracker. Lower the video playback speed to gain extra precision.

Updates¶

Various bug fixes and performance improvements

4.0.44 - March 22, 2022¶

New Features¶

Users can now specify the port number that the Redactor Server will listen on during installation.
The Windows installer now allows users to choose the installation path.
NVIDIA Ampere GPUs are now supported. (e.g. RTX 3080)
Older CPUs without AVX2 are now supported.

Updates¶

The Redactor API is now disabled by default and must be manually enabled. A proper license must be obtained first.

Bug Fixes¶

Videos with non-zero timestamps could have issues with redaction boxes not being in-sync during playback or in the exported video

4.0.43 - March 14, 2022¶

New Features¶

Redactor API users can now specify AWS_ACL_CREATE & AWS_ACL_WRITE environment variables for use with S3.

Updates¶

Initial support added for Azure Gov URLs.

Bug Fixes¶

Azure Gov URLs were failing to be recognized as valid Azure paths for API users. Note: SAS URLs are not currently supported for input/output URIs.
Chromium browsers were having issues with redaction boxes not being in-sync for certain types of videos.

4.0.42 - March 1, 2022¶

Bug Fixes¶

The "Video Load" error message was not displaying the correct title.

4.0.41 - February 17, 2022¶

Bug Fixes¶

The Desktop application could timeout during launch if more than 100 videos were in its session list. This was fixed by automatically cleaning up old entries and extending the timeout timer a bit.

Updates¶

Various logging updates were made to assist with detecting GPUs and other processes.

4.0.39 - February 10, 2022¶

Updates¶

Videos are now normalized on import to remediate negative timestamps.
For systems with an NVIDIA GPU, the computer vision models available for use are now dependent upon the amount of GPU RAM.
The thumbnail generated from a "Automatically Track" process is now taken from the same time the video player was at when the object box was drawn.
Improved thumbnail generation for very short videos.

Bug Fixes¶

Modifications to Static Tracks were not getting saved correctly.

4.0.38 - February 1, 2022¶

New Features¶

Users can now force a conversion (transcode/transmux) to occur on a video during import. This might help fix videos that have some missing packets or other issues.

Updates¶

Writing data files is now more safer and more robust.
The Sighthound Redactor logo has been updated with the new style.
The auto-conversion process for problem input videos has been improved.
The computer vision models were updated.

Bug Fixes¶

Canceling an "Export Video" process could cause "In Use" to appear on the video on the main page.
Canceling an "Automatically Redact" operation for a newly uploaded video could fail or crash.
Clicking a recently modified item in the Objects List would not correctly go to its first frame in the video.
There was a detection box synchronization issue for videos with non-zero, positive timestamps in Chromium browsers (Chrome, Edge, etc.)
Videos that were previously showing as "In Queue" would not see the message change to "Processing" when work actually started on them.

4.0.35 - October 22, 2021¶

New Features¶

Users can now specify which Redaction Mode (Standard/Enhanced) to use for auto redaction.
Objects in the Objects List can be deleted by right-clicking and selecting "Delete".
A new REDACTOR_SAVE_DIR environment variable can restrict exports to a specific folder in Redactor’s desktop version. (N/A in client/server)
A new REDACTOR_SAVE_DIR_ALLOW_SUBDIRS environment variable allows users to create subfolders in the path specified in REDACTOR_SAVE_DIR in Redactor’s desktop version. (N/A in client/server)

Bug Fixes¶

Deleting an object in the Objects List would scroll the list all the way back to the first item.
Static redaction boxes were not properly saving after they were modified.
The "Remove all audio" button required a double-click to enable. This was changed to a single click.
Multiple processes accessing a local settings/data file could cause corruption. They are now locked during update.
Docker versions of Redactor could delete the license file if the Docker network settings changed.

4.0.32 - August 9, 2021¶

Bug Fixes¶

The Audio Editor was extremely sluggish when working with longer videos.

4.0.31 - July 26, 2021¶

Bug Fixes¶

The audio redaction "tone" was having issues with the HE-AAC audio codec on export.

4.0.29 - August 9, 2021¶

Bug Fixes¶

The Redactor Server version would mistakenly restart the Windows service when thousands of objects were detected after running auto-detection.