Plant Floors
Use task-area views where the required equipment is visibly distinguishable.
A worker near machinery without visible required gear can need a supervisor’s attention. StreamVLM™ can surface the visible condition without deciding compliance from a frame alone.
The Visible Condition
The condition in scope is a person in a selected work area without the specified visible safety gear, such as a hard hat. It is a visible scene detail, not a diagnosis, a conclusion about intent, or an automatic decision.
In StreamVLM™, an authorized operator writes a prompt such as “Alert when a worker near machinery is not wearing a hard hat.” The model evaluates selected live frames against that prompt and sends a match, camera, location, and time into the IREX event workflow for human review.
StreamVLM™ runs on a GPU node in the customer instance. Each prompt has its own confidence threshold, alert cooldown, event type, and camera selection.
Explore StreamVLM Prompt Detectors →Where It Helps
Use task-area views where the required equipment is visibly distinguishable.
Set different prompts for different work zones when hard hats and other PPE rules differ.
Let supervisors review the frame before addressing a worker or recording a compliance action.
Illustrative Camera Frames
These illustrative frames show the kind of visible condition a prompt can surface. They are not real incidents, deployments, or proof of detection accuracy.
A reviewer checks the site’s PPE rule and current work status before contacting the safety team.
From Prompt to Action
Write a concise prompt for a person in a selected work area without the specified visible safety gear, such as a hard hat, then select the cameras and times where the condition warrants attention.
Open the event with its camera, location, time, and available video context. Confirm what is visible before treating the match as actionable.
The authorized team applies the local procedure. StreamVLM supplies a review cue; it does not decide the response.
Limits to Test
No. The prompt can surface missing visible gear in a defined camera view. A qualified site process determines requirements and compliance.
No. It reports a match to the visible condition in the prompt. An operator reviews the frame and local context before deciding what it means.
Start with representative cameras, define what makes a useful alert, then adjust the camera selection, confidence threshold, and alert cooldown during a camera-specific pilot.
IREX does not publish a universal accuracy figure for this prompt. Image quality, camera angle, lighting, occlusion, and the local definition of a useful alert all affect results.
Begin with one work area, one visible item of gear, and a supervisor-led review process that reflects the site’s own safety rules.