DVR-Scan v1.7

Release Notes

DVR-Scan 1.7 ships with a completely new UI, and supports faster video decoders (up to 50% better scanning performance):

The UI can be started by running dvr-scan-app, and is installed alongside the existing command line interface dvr-scan. Config settings and region files can be shared seamlessly between both. Feedback on the new UI is welcome.

Changelog

• [feature] New GUI now available across all platforms, can be launched via dvr-scan-app
  • Has UI elements for all settings, supports editing motion regions
  • Shows scan progress in real-time
• [feature] Add ability to control video decoder via input-mode config option (opencv, pyav, moviepy)
  • Allows switching between OpenCV (default), PyAV, and MoviePy for video decoding
  • Certain backends provide substantial performance benefits, up to 50% in some cases (let us know which one works best!)
• [bugfix] Fix crash on headless systems that don't have pillow installed
• [general] The region editor no longer prompts for a save path if one was already specified via the -s/--save-regions option
• [general] A size-limited logfile is now kept locally, useful for filing bug reports
  • Can be controlled with config file options save-log (default: yes), max-log-size (default: 20 kB), max-log-files (default: 4)
  • Path can be found under help entry for --logfile by running dvr-scan --help or dvr-scan-app --help
• [general] Minimum supported Python version is now 3.9

DVR-Scan v1.7 Beta 2

Release Notes

DVR-Scan 1.7 Beta 2 includes some bugfixes for the original beta release and improves error handling. To run the new GUI, launch dvr-scan-app instead of dvr-scan. The application fully integrates with the region editor, and supports loading/saving existing config files. Saved config/region files are compatible with the CLI.

Windows Download

Installer (MSI) or Portable (ZIP)

The installer will create a shortcut to the new app if requested. If you use the portable distribution, the GUI is called dvr-scan-app.exe. Both versions are compatible with existing DVR-Scan versions and can upgrade installations in-place.

Python Package (All Platforms)

Python Wheel (.whl) or Source Distribution (.tar.gz)

Install either Python distribution with pipx or pip as usual.

Changelog

• [bugfix] Fix "too many values to unpack" exception when scanning video
• [bugfix] Fix incorrect calculation of total frames when showing progress
• [bugfix] Fix drag-and-drop shortcut not working due to not parsing input paths from command line
• [general] Show alert on any errors during motion scanning

Bug Reports & Feedback

See #198 to provide feedback or submit bug reports.

Screenshots

Linux (xfce):

Windows:

DVR-Scan v1.7 Beta 1

Release Notes

DVR-Scan 1.7 Beta 1 includes a completely new GUI frontend for DVR-Scan. To run it, launch dvr-scan-app instead of dvr-scan. The application fully integrates with the region editor, and supports loading/saving existing config files. Saved config/region files are compatible with the CLI.

Windows Download

Installer (MSI) or Portable (ZIP)

If you use the portable distribution, the GUI is called dvr-scan-app.exe.

Python Package (All Platforms)

Python Wheel (.whl) or Source Distribution (.tar.gz)

Install either Python distribution with pipx or pip as usual.

Bug Reports & Feedback

See #198 to provide feedback or submit bug reports.

Screenshots

Linux (xfce):

Windows:

DVR-Scan v1.6.2

Release Notes

DVR-Scan 1.6.2 includes a completely reworked region editor that should provide a much more consistent UI experience across all platforms.

Changelog

• [feature] Vastly improved region editor UI:
  • Reimplemented entire UI with Tcl/Tk to provide much more consistent experience across platforms
  • All options now have dedicated UI controls in addition to keyboard shortcuts
  • System-specific shortcuts like undo/redo now work as expected
  • Various performance and usability improvements including zoom/pan and shape selection
• [feature] Add new variance-threshold config option to control how MOG2 controls which pixels are described by the current background model #163
• [feature] Add new --thumbnails option to extract images from each event, use --thumbnails highscore to save frame with highest motion score #159
• [feature] Add --use-pts option to allow using presentation time instead of frame number for timestamps #170
• [bugfix] Fix incorrect framerate detection in Windows builds #174
• [general] Updates to Windows distributions:
  • Python 3.9 -> Python 3.12
  • OpenCV 4.10.0.82 -> 4.10.0.84
  • Ffmpeg 6.0 -> 7.1
• [general] Built HTML documentation is now included with all distributions

DVR-Scan v1.6.1

Release Notes

DVR-Scan 1.6.1 includes some important fixes and improvements for the region editor. Minimum supported Python version is now 3.8.

Changelog

• [bugfix] Corrupt frames are no longer encoded and are skipped when encoding #151
• [bugfix] Fix region-editor config file option having no effect #154
• [bugfix] The region editor now always prompts on any unsaved changes #161
• [general] The region editor no longer prompts for a save path if -s/--save-region was specified
• [general] Regions added via command line are now be merged with those loaded from the load-region config option
• [general] Add new learning-rate config option to control how fast the background model is updated #158 (thanks @felixscheffer)
  • Value between 0.0 and 1.0 controls how much weight is placed on next frame, or -1 for auto
  • 0.0 implies no update to the model, 1.0 will re-initialize it completely on each frame
  • Default value of -1 is for automatic, which is unchanged from previous releases

DVR-Scan v1.6

DVR-Scan 1.6 greatly improves masking capabilities with the new region editor (user guide):

Multiple regions can now be defined with any shape, size, and complexity. Region data can be saved to a file and loaded again. Regions can also be specified by command line.

There are also several other bugfixes and improvements, such as improved seeking performance.

Changelog

• [feature] New region editor -r/--region-editor allows creation of multiple regions without shape restrictions, replaces -roi/--region-of-interest
• [feature] Multiple regions of interest (rectangular or polygonal) can now be created:
  • Using the new region editor by adding the -r/--region-editor flag: dvr-scan -i video.mp4 -r
  • New -a/--add-region replaces -roi/--region-of-interest option: dvr-scan -i video.mp4 -a 5 5 20 5 20 20 5 20
  • Regions can now be saved to a file: press S in the region editor or use -s/--save-region
  • Regions can now be loaded from a file: press O in the region editor or use -R/--load-region
  • Config files can specify a region file to use by default with the load-region option, replaces the region-of-interest setting
• [feature] New -fm / --frame-metrics option draws motion score on each frame to help tune detection parameters
• [cli] Short flag -v is now used for --verbosity, replaced by -V for --version
• [cli] -roi/--region-of-interest is now deprecated, replaced by region editor and add/save/load region flags
  • Specifying this option will display the ROI in the new region format allowing you to update usages more easily
• [general] Improved seeking performance, using -st/--start-time is now much faster (#92)
  -detection-parameters)
• [general] Noise reduction kernel can now be disabled by setting -k/--kernel-size to 0 (#123)
• [general] Include stack traces in logfiles when setting --verbosity debug
• [bugfix] Add max-score option to config file to fix CNT mode always treating first few frame as motion, default is 255.0 #119
• [bugfix] Fix timecode format HH:MM:SS[.nnn] being rejected for start/end time (#141)
• [bugfix] Fix incorrect RGB mapping for config file (values were treated as BGR instead)
• [other] Config option timecode has been renamed to time-code to match the command-line option
• [other] Config options that started with timecode- have been renamed to start with text-, and are now shared between the time-code and frame-metrics overlays:
  • time-code-margin is now text-margin
  • time-code-font-scale is now text-font-scale
  • time-code-font-thickness is now text-font-thickness
  • time-code-font-color is now text-font-color
  • time-code-bg-color is now text-bg-color

Known Issues

• Some prebuilt archives include documentation which references the load-region config option with the incorrect name (region-file) #153

DVR-Scan v1.5.1

Minor re-release of v1.5 to fix crash when opening multiple videos.

Changelog

• [bugfix] Fix crash when opening multiple input videos (#95)
• [bugfix] Fix incorrect warning regarding frame decode failures at end of video

DVR-Scan v1.5

Release Notes

• Significant performance improvements on multicore systems
• Support wildcards/globs as inputs for scanning entire folders (-i folder/*.mp4)
• Allow use of ffmpeg for better output quality (-m ffmpeg) or codec-copying mode (-m copy)
• Configuration files are now supported, see documentation for details
  • Can specify config file path with -c/--config, or create a dvr-scan.cfg file in your user config folder
• Windows binaries are now signed, thanks SignPath.io (certificate by SignPath Foundation)
• Experimental Nvidia CUDA® support has been added (set -b MOG2_CUDA)
  • If installing via pip, requires manual installation of OpenCV compiled with CUDA support
  • If downloading Windows version, make sure to download the GPU-enabled build (dvr-scan-1.5-win64-cuda.zip)
  • CUDA-enabled builds only support GTX 900-series and above GPUs, do not include the CNT method, and are not yet code signed
• Minimum supported Python version is now 3.7
• Minimum supported OpenCV version is now 3.x

Command-Line Changes

• New options:
  • -c/--config - set path to config file
  • -d/--output-dir - set directory to write output files (default is working directory)
  • -m/--output-mode - set output mode (one of: opencv, ffmpeg, copy)
  • -mo/--mask-output - path to write motion mask for analysis
  • --verbosity, --logfile - control output verbosity, path to save log output
• -i/--input now supports globs/wildcards to scan entire folders, e.g. -i folder/*.mp4
• Change default value for -l/--min-event-length to 0.1 seconds, previously was 2 frames
• Long form of -roi has been renamed to --region-of-interest (previously was --rectangle-of-interest)
• -c is now used for --config, previously was for --codec
• Add experimental MOG2_CUDA option for -b/--bg-subtractor
• Rename existing MOG option to MOG2
• --codec has been removed, the value should now be set using config file option opencv-fourcc

Changelog

• [feature] Configuration file support and new -c/--config argument to specify path to config files (#77)
  • Breaks existing behavior of -c (was previously the shortform of --codec)
• [feature] Add support for multiple output modes via -m/--output-mode argument (#27, #42)
• [feature] Experimental support for GPU-based CUDA MOG2 filter (#12)
• [feature] Video encoding and decoding are now done in parallel with the scanning logic leading to improved performance on most systems (#52)
• [feature] Add support for exporting motion masks via -mo/--mask-output argument
  • Useful for detailed analysis or tuning of detection parameters
  • ffmpeg can be used to generate output videos by specifying -m ffmpeg
  • Codec-copy mode, using ffmpeg, can be used by specifying -m copy
• [feature] Add --verbosity and --logfile arguments to provide more control over program output
• [feature] Allow scanning entire folders using wildcards with -i/--input (#5)
  • Glob expansion is also performed on each input path directly, so quoted globs also function correctly
• [bugfix] Fix incorrect results when -st/--start-time is set
• [bugfix] Event start times are now correctly calculated when using -fs/--frame-skip (#68, #70)
  • Note that all skipped frames within the event window are included in motion event, thus the calculated start time may be slightly earlier
• [bugfix] Only get screen resolution when required (was causing issues on headless machines)
• [bugfix] Fix output messages conflicting with progress bar shown during scanning
• [bugfix] Output events now start from 1 to align with the event list
• [bugfix] Event end times now include the presentation duration of the last frame
• [bugfix] Small values for -l/--min-event-length are now handled correctly, previously would cause an error
• [enhancement] Progress bar now indicates how many events have been detected so far
• [enhancement] Change default value for min_event_len to 0.1 seconds, previously was 2 frames (#91)

Known Issues

• Attempting to open multiple input videos will cause DVR-Scan to crash, fixed in v1.5.1 (#95)
• Variable framerate videos (VFR) are not fully supported, and will yield incorrect timestamps (#20)
• Video output when using frame skip and -m opencv (default output mode) will result in frames missing from the exported videos (#81)
  • Use -m ffmpeg or -m copy as a workaround
• Multiple input videos are not supported with -m ffmpeg or -m copy (#86)
  • Use ffmpeg to concatenate/merge input videos before processing as a workaround
• CUDA builds do not include the CNT option for -b/--bg-subtractor

DVR-Scan v1.5-beta

Documentation: DVR-Scan v1.5-beta Documentation

Install from pip via:

pip install dvr-scan[opencv] --pre --upgrade

Windows builds (regular and with experimental GPU support) are available below.

Release Notes

• Significant performance improvements on multicore systems
• Support wildcards/globs as inputs for scanning entire folders (-i folder/*.mp4)
• Allow use of ffmpeg for better output quality (-m ffmpeg) or codec-copying mode (-m copy)
• Configuration files are now supported, see documentation for details
  • Can specify config file path with -c/--config, or create a dvr-scan.cfg file in your user config folder
• Windows binaries are now signed, thanks SignPath.io (certificate by SignPath Foundation)
  • Shout-out to Paul from SignPath, you were awesome and a huge help in making this happen
• Experimental CUDA® support has been added (set -b mog_cuda)
  • If installing via Pip, requires manual installation of OpenCV compiled with CUDA support
  • If downloading Windows version, make sure to download the GPU-enabled build (dvr-scan-1.5.dev1-win64-cuda.zip)
  • CUDA-enabled builds are not signed and should be considered experimental
• Minimum supported Python version is now 3.7
• Minimum supported OpenCV version is now 3.x

Changelog

Command-Line Interface

• New features/arguments (see below for more details):
  • -c/--config - set path to config file
  • -d/--output-dir - set directory to write output files (default is working directory)
  • -m/--output-mode - set output mode (one of: opencv, ffmpeg, copy)
  • -mo/--mask-output - path to write motion mask for analysis
  • --verbosity and --logfile - control output verbosity and path to save output
• -i/--input now supports globs/wildcards to scan entire folders, e.g. -i folder/*.mp4
• Long form of -roi has been renamed to --region-of-interest (previously was --rectangle-of-interest)
• -c is now used for --config, previously was for --codec
• Add experimental mog_cuda option for -b/--bg-subtractor
• --codec has been removed, the value should now be set using a config file

General

• [feature] Configuration file support and new -c/--config argument to specify path to config files (#77)
  • Breaks existing behavior of -c (was previously the shortform of --codec)
• [feature] Add support for multiple output modes via -m/--output-mode argument (#27, #42)
• [feature] Experimental support for GPU-based CUDA MOG2 filter (#12)
• [feature] Video encoding and decoding are now done in parallel with the scanning logic leading to improved performance on most systems (#52)
• [feature] Add support for exporting motion masks via -mo/--mask-output argument
  • Useful for detailed analysis or tuning of detection parameters
  • ffmpeg can be used to generate output videos by specifying -m ffmpeg
  • Codec-copy mode, using ffmpeg, can be used by specifying -m copy
• [feature] Add --verbosity and --logfile arguments to provide more control over program output
• [feature] Allow scanning entire folders using wildcards with -i/--input (#5)
  • Glob expansion is also performed on each input path directly, so quoted globs also function correctly
• [bugfix] Fix incorrect results when -st/--start-time is set
• [bugfix] Event start times are now correctly calculated when using -fs/--frame-skip (#68, #70)
  • Note that all skipped frames within the event window are included in motion event, thus the calculated start time may be slightly earlier
• [bugfix] Only get screen resolution when required
• [api] Remove dvr_scan.timecode and FrameTimecode, replace with scenedetect.FrameTimecode

Known Issues

• Variable framerate videos (VFR) are not fully supported, and will yield incorrect timestamps (#20)
• Video output when using frame skip and -m opencv (default output mode) will result in frames missing from the exported videos (#81, can use -m ffmpeg or -m copy as a workaround)
• Multiple input videos are not supported yet when using -m ffmpeg or -m copy (#86)
• Fixed in latest beta: Crash with timecode framenumber must be greater than zero (#90)

Code Signing Policy

Windows EXE/MSI Builds: Free code signing provided by SignPath.io, certificate by SignPath Foundation.

DVR-Scan v1.4.1

Release Notes

This release includes fixes for incorrect event start/end times when using frame skipping, and improvements to the bounding box overlays.

Important: Aside from critical bugfixes, DVR-Scan 1.4 is the last version supporting Python 2.7. Starting from v1.5, the new minimum supported Python version will be 3.6 3.7 (see #83 for details).

Python users can install via pip:

pip install --upgrade dvr-scan[opencv]

Builds for Windows (installer + portable .zip) are available below.

Changelog

• [bugfix] Event start times are now correctly calculated when using -fs/--frame-skip (#68, #70)
  • Note that event start/end times may still be off by how many frames are skipped due to loss of context with respect to frame-accurate motion detection
• [bugfix] Event end times now take into account the number of skipped frames if -fs/--frame-skip if specified (e.g. it is assumed all skipped frames contained motion)
• [bugfix] Bounding box smoothing now takes into account -fs/--frame-skip (#31)
• [bugfix] Bounding boxes now cover all frames with motion (#31)
  • In v1.4 only the frames after -l/--min-event-length frames had passed were covered

Known Issues

• Variable framerate videos (VFR) are not fully supported, and will yield incorrect timestamps (#20)
• Using -st/--start-time will yield incorrect results