SMT_horizontal_white (1).png

SMT’s Virtual Fans, Ads and Brands (V-FAB) system has been designed to minimize the effect of empty stadiums and arenas that sports leagues are facing as sports resume without spectators. V-FAB reintroduces the spectacle and energy inherent in every live sporting event: cheering fans rooting on their favorite teams.

V-Fab_NHL_Slide 1.jpg
V-Fab_NHL_Slide 3.jpg

SMT’s V-FAB virtual insertion system creates a context-synchronized, visually pleasing atmosphere of a match played in a full arena.

SMT’s V-FAB combines:

  1. A LiDAR-scanned 3D model of an arena

  2. A finite set of select game cameras

  3. Team-branded animating fan graphics

  4. Animating ads

  5. Animating brand banners

  6. Live social media fan reaction images

  7. SMT’s proprietary virtual insertion and AI technology

lidar mls.png

V-FAB optionally supports the parallel delivery of geo-targeted ad content

V-Fab_NHL_Slide 2.jpg
V-Fab_NHL_Slide Stella.jpg
V-Fab_NHL_Slide Heinekin.jpg

Zoned virtual ads can be geographically targeted to multiple regions. Ads can be targeted on a per-market basis for local, national and international broadcasts and are excellent options for companies to build global brand awareness. 

V-FAB optionally supports inserting live fan reactions into the arena

V-Fab_NHL_Slide ZOOM Jumbo.jpg
V-Fab_NHL_Slide ZOOM Fans.jpg

Panels of live fan social media reactions are inserted onto virtual video screens or onto fan-branded sections.

V-FAB operator tasks and responsibilities

Production truck.jpg

During the game

  1. Oversees the running of insertion scripts onto Scene Target Areas

  2. Inserts manually triggered content based on production cues and game context

Prior to first use

  1. Ingests the 3D model, chooses the V-FAB camera set, and enumerates the set of Scene Target Areas

  2. Ingests and maintains the art asset library

a. Team-related assets (i.e., logos, artwork, color palettes)

b. Static sponsor/brand signage

c. Animating ads, brands, promos, mp4s

Prior to game on Game Day

  1. Build scripted playlists

  2. Select home and visitor team palettes and art assets

  3. Add any game-specific art assets, sponsor art, etc.

  4. Configure all V-FAB Camera locations

V-FAB camera section guidelines

SMT’s V-FAB system uses computer vision and AI processing to:

  1. Recognize and track visible fiducials in a V-FAB camera's field-of-view; and,

  2. Constantly calculate the V-FAB camera’s position in 3D space relative to the changing image in each new video frame.


A “fiducial” is a known point in real space that a V-FAB camera uses as an image-processing reference point and visual anchor. Examples include painted lines on the pitch grass, or 3D scanned and modeled areas of the arena.


The V-FAB rendering application uses the V-FAB camera’s POV tracking data, the 3D model of the venue, and a library of user-selected content to virtually insert video and animating graphical content into the video scene.


Cameras that will run V-FAB during a broadcast are chosen by the broadcast producer from the set of existing cameras covering that sporting event.  Some cameras are perfect for V-FAB; however, not all on-site cameras can run or should attempt to run V-FAB.  Important limitations must be understood by the broadcast producer.

V-FAB camera specifications, recommendations, limitations

V-FAB camera selection 

  1. V-FAB cameras must be able to “see” fiducials, (i.e., lines on the pitch and/or surveyed arena locations) in each video frame to track properly; ergo, not all camera POVs are suited for V-FAB.

  2. Primary game-action cameras with wide fields of view such as the main game and left/right offside camera locations are well suited and perfect candidates for V-FAB.

  3. Closeup and tight ISO cameras, high-zoom level coverage cameras, and handheld sideline cameras are not good options for V-FAB because they do not show sufficient fiducials in their fields of view.  Recommendations for choosing V-FAB cameras are:

a. No ISO cameras

b. No GLT (Goal Line Technology) cameras

c. No tight-zooming cameras; lock-offs are OK if enough fiducials are in the field of view.

V-FAB camera location

  1. A camera's physical location in the environment (relative to the pitch) is a key determinant of V-FAB's effectiveness.

a. The height and distance of the camera from the field/pitch are important considerations.

b. The higher the camera, and the farther away from the near sideline, the better.

c. Sufficient height/distance is needed to provide views of critical field-level fiducials, e.g., the near sideline.

V-FAB camera operation

  1. Wider framing allows the V-FAB camera to see more fiducials in its field-of-view, facilitating accurate camera tracking.

  2. Narrow framing in ISO shots can lead to loss of camera tracking. WIDER IS BETTER.

V-FAB camera Game Day calibration (3-4 hours before kickoff)

  1. V-FAB camera operators must shoot a series of calibration clips from their POV (3-5 mins).

  2. Each calibration clip is a series of slow, sweeping pans of the scene at various zoom levels.

  3. The V-FAB system records the calibration clips shot from each V-FAB camera’s POV.

  4. The V-FAB operator calibrates each V-FAB camera using its calibration clip (20-30 mins each).

  5. SMT recommends allowing a 90-minute buffer between the end of calibration and the kickoff.

How many V-FAB cameras can the V-FAB system support?

There is no theoretical limit to the number of V-FAB cameras that can be supported by the V-FAB system; however, there are very real practical limits based on:

  1. The limited and finite set of cameras on-site that are suitable to run V-FAB; and,

  2. The amount of time needed for the V-FAB operator to calibrate the set of V-FAB cameras on Game Day prior to match start. In practice, three game cameras with optional fourth or fifth is norm.

V-FAB practical questions

What is the process for getting V-FAB in my facility?

  1. Create or acquire a LiDAR scan of the facility that is to be modeled in 3D.

  2. Choose the V-FAB configuration (i.e., downstream, on-site, REMI) for your TV production environment.

  3. Choose how many, and which, V-FAB cameras will be supported in each arena.

  4. Hire SMT to build the 3D model from the LiDAR scan.

  5. License the V-FAB system from SMT.

  6. Hire SMT to install and test V-FAB in your sport facility or central TV production studio.

  7. Configure/load V-FAB with your own art asset library.

  8. Hire SMT to operate V-FAB or hire SMT to train your operator. 

What does V-FAB cost?

  1. V-FAB is licensed annually on a per-game basis pricing model with rates determined by number of arenas, configuration, camera count, and the system options selected.

How soon can I get V-FAB operational?

  1. As a general guideline, V-FAB can be running in location within three weeks of SMT receiving the LiDAR scan.