logo logo

Investigation Dashboard

The Investigation dashboard allows you to process video files and analyze the results as a list of detected a faces and identified persons.

The system leverages the power of Rapid Video Processing, which enables you to process a list of video files faster than real-time. You have the flexibility to fully configure all processing parameters before initiating the process.

Investigation Dashboard

⚠️ In the current version, Rapid Video Processing supports face detection and face identification, but it does not support pedestrian detection.

Investigation Dashboard Sections

The main screen of the Investigation dashboard, for any investigation use-case, displays the video list. The Investigation dashboard enables you to manage video files, process them, and review the results for each video file.

The Investigation dashboard consists of:

  1. Title bar
  2. Selected filters area
  3. Video list
  4. Full-text search by name
  5. All available filters

The Investigation Dashboard - sections

Title bar

Displays the page title and actual number of uploaded video records. Use the + ADD VIDEO button on the right to open the upload page.

Selected filters area

If you select any filter from the list of available filters, it will appear in the selected filters area. To remove one filter, click on X button. To remove all selected filters, click on CLEAR ALL button.

Video list

Displays all uploaded video records with basic information and actual processing status. Click on any of them to open the detail of the video record.

Full-text search

To perform a full-text search by name for video records , type text in the search box and press Enter.

All available filters

The SmartFace Station allows you to use four filters to fine-tune your video search.

  • Tags
    Use tags to categorize and organize your video files. Tags are keywords or labels that you can assign to videos to make them easier to find later. You can add one or more tags to your video files, allowing you to filter and search for specific videos based on these tags.

  • Status
    The status filter allows you to filter videos based on their current state. It can help you quickly identify which videos are Ready to process, Processing, Processed or in Processing error state.

  • Recorded at
    This filter enables you to narrow down your video search by specifying the date and time when the video was recorded.

  • Uploaded at
    The uploaded at filter allows you to filter videos based on the date and time they were uploaded to the SmartFace

Video investigation process

To do the investigation on available video files, please follow these steps:

Click on the links below to see details of each step:

  1. Upload video files
  2. Process video records
  3. Do the investigation

See the diagram for the video record lifecycle and possible video record statuses:

StatusDescriptionPossible action
READY TO PROCESSVideo record has been uploaded.Video record can be processed.
PROCESSINGProcessing has started, but not finished.No action can be taken.
PROCESSEDVideo record has been processed successfully.Video record can be reprocessed. Results of processing can be analyzed.
PROCESSING ERRORError occurred during video record processing.Video record can be reprocessed.
ℹ️ Please note that you can edit video record, manage tags and download at any time. Deleting the video is not possible only during the ongoing processing.

Video files upload

A user can upload one or more video files from his computer using the video upload screen. To open video upload screen, click on + ADD VIDEO button on Investigation video list page.

Investigation video list screen - ADD VIDEO button

To add new video records, follow these steps:

  1. Drag and drop video files to the designed area, or select the files from the file system by clicking the choose videos link

Video upload page - initial state

  1. Add tags to all video records what will be created (optional).

Video upload page - added videos for upload and tags

  1. Click UPLOAD

Notes:

  • It is possible to upload multiple video files at the same time

  • You can see progress of uploading

  • Successfully uploaded video files are marked with a green check mark

  • During the upload process, it is not possible to add or delete tags

  • You can cancel uploading anytime

  • A red exclamation mark will appear in case of an upload error. You can retry uploading, or you can delete video files in error status

Video upload page - uploading process

Video upload page - uploading process was successful

Video record processing

To detect and identify faces in uploaded video files, video records must be processed. You can start processing on any video record in READY TO PROCESS state by clicking PROCESS button on the video record detail tab.

Video record detail tab - READY TO PROCESS status

A new tab with processing configuration will appear where you can change processing attributes. To start the processing, click on START PROCESSING button.

Processing configuration tab

After successful processing, video record is in PROCESSED state and all the data can be presented on history page. You can show processing results by clicking on SHOW IN HISTORY button, on video detail tab.

Processing attributes configuration

By changing processing attributes you can influence what faces will be detected, how many identifications will be done, how log will system process the video and many more. See attributes detailed information in the table below:

Attribute nameAttribute DescriptionDefault valueValue interval
VIDEO STARTStart time when the video was recorded.If the time of video start is not provided, SmartFace displays the time when the video record was uploaded.
COARSE GRAIN INTERVALThe period in which SmartFace makes a detection on a frame from the video.500 ms500 - 2000 ms
FINE GRAIN INTERVALThe period in which SmartFace perform more frequent detections around the time of the detected face.500 ms100 - 500 ms
MIN FACE SIZEMinimum size of the face what system can detect.35 px
MAX FACE SIZEMaximum size of the face what system will detect.600 px
MAX TRACKING FACESThe maximum number of faces detected on one frame.20
FACE SAVE STRATEGY

The strategy for saving data from all face detections, you can choose from:

  • All- All faces for the detected person are stored.
  • FirstFace - Only the first detected face of a person is stored.
  • BestFace - Only the best face for the detected person is stored. Which face is the best is decided when the person is lost from tracking.
  • FirstFace, BestFace - The first and the best face for the detected person are stored.
  • MatchedOnly - Only faces which are matched against a watchlist member are stored.

⚠️Set this wisely. Strategy to save all detected objects can generate huge amount of data.

FirstFace, BestFace
FACE CONFIDENCE THRESHOLDThe confidence threshold for the face detection4500 - 10 000
MATCH RESULTS COUNTHow many matches can be provided for one face1
FULL FRAME SAVINGEnables the storage of fullframes.On/OffOn

Reprocessing video record

You can process the video record again when you see that the results are not good enough, clicking on the REPROCESS button on the video record detail tab.

ℹ️ You can also proceed with REPROCESS when your video record is in PROCESSING ERROR state.

Video record analysis

SmartFace Station can display face detection and identification results for each processed video record. To show results in standard view of history page:

  1. Click on INVESTIGATION icon in the left menu
  2. Click on the video record to display video record detail
  3. Click on SHOW HISTORY button

History event page for video investigation purposes

The main screen of history events consists of:

  1. page title part with the number of results displayed and the refresh button
  2. selected filters area - in this part, the video you are analyzing is preselected
  3. filtered results list
  4. full-text search and all available filters

The SmartFace Station allows you to use filters to fine-tune your event history search. For more information, click here.

People identification in the video

You can add people for identification to the watchlist and start the processing - use this approach when you need to identify more people on many video files. Use face search function on history page - use this approach when you need to find one person on one or many video files.

Video record detail

A user can open video record detail by clicking on any video record on Investigation video list screen.

Investigation dashboard - video detail

Video record detail, depending on the actual state, can contain different set of information. A complete list of information and its description can be found below:

InformationDescrption
NameName of the video record.
DescriptionDescription of the video record.
Status

  • Ready to process
  • Processing
  • Processed
  • Processing error

see video record status description here

TagsTag management component. You can create a new tag or assign an existing tag to the video record.
Time attributes

  • Uploaded at - the time when the video record was uploaded.
  • Processing - the time when the video was recorded (start).
  • Processed - the time when the video was recorded (end).
  • Processing error - the duration of the video record.

StatisticsThe SmarrtFace Station displays basic video statistics to the user after successful video processing.

  • Tracked faces - the number of unidentified faces.
  • Identified persons - the number of all identifications.
  • Unique identified persons -the number of unique identified persons.

Button section

  • Process or Reprocess - process or reprocess selected video record.
  • Show in history - show results of the successful video processing on history event page.
  • Download - download video record.
  • Delete - delete video record and all assigned detection and identifications.

Processing attributesProcessing attributes which were use during the video record processing.

Variations of video record detail

The information in the video record detail varies according to video record status:

Video detailStatus and description

  • Video record is ready to process.
  • You can edit name, description and tags.
  • The video start time is by default the same as the time when the video record was uploaded.
  • The video end time and the duration is not available yet.
  • You can start the process by clicking the button PROCESS button.

  • Ongoing video processing.
  • You can edit name, description and tags.
  • Video detail displays processing attributes.

  • Video record was successfully processed.
  • You can edit name, description and tags.
  • It is possible to reprocess video record with new parameters by clicking REPROCESS button.
  • Click on SHOW IN HISTORY button to see results of the video processing.

  • Video record processing was not successful.
  • You can edit name, description and tags.
  • The video end time and the duration are not available.
  • You can try to reprocess, download or delete video record.
  • It is not possible to show results of the video processing.

Investigation module configuration

SmatFace Station v5_1.12 introduced a new investigation module, where user can upload video files for investigation. This functionality requires configuration changes for proper connection between the Min.IO and the SmartFace Station.

It is necessary to set these parameters in the configuration file in SmartFace Station:

  • S3_ENDPOINT - machine host name or IP where MIN.IO/S3 is installed (default - pseudo IP address of the local computer)

  • S3_ACCESS_KEY - login

  • S3_SECRET_KEY - password

  • S3_BUCKET - a container for objects stored in S3.

Default values after SmartFace Station installation are:

S3_ENDPOINT=http://127.0.0.1:9000     
S3_ACCESS_KEY=minioadmin
S3_SECRET_KEY=minioadmin
S3_BUCKET=inno-smartface

This will not work, when you run SmartFace Station on a client computer. The SmartFace Station will search for MIN.IO database on the client machine, not on the server.

Example with host name:

S3_ENDPOINT=http://sfit-sface-3w:9000     
S3_ACCESS_KEY=minioadmin
S3_SECRET_KEY=minioadmin
S3_BUCKET=inno-smartface

If the client computer is in the network, where sfit-sface-3w server is reachable, the user will be able to upload video files.

SmartFace Station configuration file location

Windows deployment:

The default location for .env file is: C:\Program Files (x86)\SmartFace Station\

ℹ️ Do not forget to restart the SmartFace Station service after changing the configuration.

For LINUX deployment:

The default location for .env.sfstation files is: smartface/sf-docker/

ℹ️ Do not forget to restart the SmartFace Station docker container after changing the configuration.