AI-powered redaction for video, images, and audio

New Features

• Out-of-Disk-Space Protection
  When disk usage on the app data folder exceeds the threshold, the server replies with HTTP 503 to all requests except /health and resumes automatically once it drops. Configurable via REDACTOR_OODSTRACK_THRESHOLD_PCT (default 90%) and REDACTOR_OODSTRACK_INTERVALS.
• Memory Diagnostics
  A new tracker samples heap usage and logs new high-water marks, and can write a V8 heap snapshot when usage exceeds a configurable percentage of the heap limit. Controlled via REDACTOR_DIAG_OOMTRACK_INTVL, REDACTOR_DIAG_OOMTRACK_DUMP_PCT, and REDACTOR_DIAG_OOMTRACK_DUMP_PATH.
• Optional Persisted Redaction History
  Per-session history persistence is now opt-in via the new REDACTOR_HISTORY_ENABLED env var (default 0=off), reducing disk I/O for high-throughput API workloads. For in-depth audit logs you might want to set it to 1.

Updates

• Server Stability Internal optimizations got added to make the backend more robust and responsive under concurrent API (and sometimes regular user) load.
• Operational Logging
  Active session and operation counts are now logged on each change, and OS resource limits are logged at startup.

Bug Fixes

• KeepAlive Interval with Polling Transport
  Setting REDACTOR_FORCE_POLLING_TRANSPORT=1 could hit the server with too many keepAlive POSTs. This is now back to regular, recommended intervals.
• Box Border Visuals
  The borders for background / keep-unredacted boxes did not render properly. Their original visual concept as got improved for better visibility.
• Reaction to Malformed Requests
  All effective endpoint flows are now validated thoroughly againt malformed traffic (observed on publicly-exposed/non-firewalled hosts)

‍

• Smart Fill redaction mode: new dedicated option in the export UI for visually pleasing content-aware fill
• Improved audio gap filling: proper resampling/silence padding during conversion
• Unknown serials now show the correct rejection reason, mixed activation responses yield success
• Context menus suppressed during object creation

• GPU environment variables now configured before CUDA initialization to prevent conflicts.
• License is now validated before processing, surfacing issues immediately.
• Bulk selection context menu now shows type-specific labels.
• Deinterlace filter only applied when explicitly requested.
• Error dialogs now display more readable messages .
• Fixed crash on videos due to odd frame widths.
• Fixed video list appearing empty after refreshing the browser.

Stability Improvements

• Fixed a crash that could occur while probing video decoding capabilities on certain hardware and CUDA driver combinations. Decoder probing is now disabled to prevent unintended crashes.
• Resolved a race condition during track resizing that could cause objects to lose their associated tracks and trigger UI errors. Existing data inconsistencies are automatically mitigated.
• Restored missing vehicle and screen detection labels in the video list.

This update focuses on improving system stability and correcting UI inconsistencies in early 7.x releases.

Expanded Detection Capabilities

Redactor now detects additional object types:

• Documents
• IDs (e.g., driver licenses, passports)
• Screens (laptops, monitors, phones)

These detection classes are also available through the API.

Modern Video Format Support

• H.265 (HEVC) video files can now be imported.
• Exports continue to use H.264 for maximum playback compatibility.

Editing Workflow Improvements

• Full undo and redo support across video boxes, audio regions, and object grouping.
• New background (reverse) redaction options allow full-scene redaction while keeping selected objects visible.
• Escape key support for quick cancellation.
• Improved high-contrast object boxes for better visibility across complex scenes.

Audit and Traceability Enhancements

• Generate downloadable audit logs with structured session activity.
• Exportable session history data for deeper review and diagnostics.
• User names displayed in the server video list when files are in use.
• Automatic hash generation for imported and exported media files.
• API session metadata snapshots available for operational insight.
• TypeScript definitions published for API integrations.

Reliability and Operational Updates

• Improved crash diagnostics and export handling.
• Clearer behavior when sessions contain no redactions.
• Improved handling of live recording sessions.
• Multiple fixes across audio timelines, export status, imports, and full-frame redaction.

Operational Controls

• Added a Data Drop flag for transient or API-driven deployments where persistent data is not required.
• Enforced immutable session identifiers to prevent conflicts when reloading stored sessions.

Stability Improvements

• Fixed backend crashes during certain storage cleanup scenarios.
• Improved audio handling when video files contain partial audio segments.

API Enhancements

• OpenAPI specification exposure for improved API clarity and validation.
• Optional request and response validation.

Infrastructure Improvements

• Automatic CPU fallback if GPU detection fails.
• Improved server reconnection handling for unstable network environments.

v6.5.1 and Earlier

Previous releases have included:

• Bulk object selection improvements.
• GPU behavior refinements for greater stability.
• Object Timeline View for precise review.
• Object grouping for complex scenes.
• Advanced head detection (including rear views).
• Folder management and smart search.
• Audio-only redaction support.
• Expanded keyboard shortcuts.
• Expanded language support.

For full technical release notes, including detailed environment variables and API configuration options, please refer to the developer documentation.