What breaks first.
Service Catalog
Five capability layers, eighteen services.
Organized the way video problems actually surface. Find and follow what’s on
camera, understand what’s happening, make it searchable in plain language, catch
what breaks the pattern, then turn it into operational metrics.
1.1 Object Detection
See what's there, without watching every frame.
Detect and locate people, vehicles, bags, equipment and other objects across video streams with high visual accuracy. Identify what appears, where it appears and when it enters or leaves the scene.
Deliverables: Object counts, object labels, confidence scores
Result: Faster search, clearer visibility, less manual review.
1.2 Cross-Camera Tracking
Follow the object, even when the camera changes.
Track people and objects across multiple cameras while maintaining their identity throughout the scene. Connect movement across locations and timelines to reconstruct where an object went and when.
Deliverables: Track IDs, movement paths, cross-camera links
Result: Continuous visibility across disconnected views.
2.1 Scene Classification
Know the scene before you analyze it.
Classify video by location, environment, activity and scene type to understand the context surrounding an event. Automatically organize footage so relevant content can be found without reviewing everything.
Deliverables: Scene labels, confidence scores, metadata
Result: Faster discovery, better context, smarter filtering.
2.2 PPE & Safety Compliance
Know when safety stops being followed.
Detect PPE usage and identify safety compliance across people, zones and activities. Flag missing helmets, masks, vests and other required equipment before non-compliance becomes an incident.
Deliverables: Compliance status, violations, alerts
Result: Safer operations, faster intervention, stronger compliance.
2.3 Behavioral Recognition
Recognize actions, not just appearances.
Identify behaviors and activities as they happen across video, from routine actions to unusual interactions. Understand what people are doing, how behavior changes and where intervention may be required.
Deliverables: Action labels, pose estimation, activity events.
Result: Action-level visibility into what's happening.
2.4 Event Recognition
Detect the event, not just the objects inside it.
Identify meaningful events such as fire, smoke, intrusion, falls and other defined activities. Combine visual signals over time to distinguish an actual event from an isolated frame.
Deliverables: Event type, time, location, severity.
Result: Earlier detection, fewer missed events, faster response.
3.1 Semantic Search
Search for meaning, not just keywords.
Find people, objects, actions and scenes using natural-language descriptions of what you're looking for. Search across footage by visual context rather than relying on filenames, timestamps or manual tagging.
Deliverables: Search results, matching frames, contextual metadata.
Result: Less scrubbing, faster answers.
3.2 Forensic Timeline Search
Go from a question to the moment it happened.
Retrieve incidents, events or objects across a defined time window and trace their occurrence quickly. Narrow long recordings into the moments that matter, with searchable timelines and bookmarks.
Deliverables: Event timelines, clips, bookmarks.
Result: EFaster investigation, clearer evidence trails.
3.3 Evidence Management
Keep the evidence connected to the story.
Organize video, events, clips and supporting metadata so every finding can be traced back to its source. Preserve relevant footage and context without losing the chain between an event and its evidence.
Deliverables: Evidence clips, metadata, reports, audit logs.
Result: Defensible evidence, easier investigation, cleaner records..
3.4 Query Analytics
Understand what your video intelligence is being asked.
Track search patterns, recurring queries and frequently investigated events across your video environment. Use those patterns to understand where teams spend time and which questions matter most.
Deliverables: Search metrics, query trends, usage insights
Result: Better workflows, sharper priorities, measurable usage.
4.1 Rule-Based Anomaly Detection
Decide what shouldn't happen, then detect it.
Define rules around movement, zones, objects, timings and activities to identify known deviations from expected behavior. Automatically flag events that match the conditions you've already defined.
Deliverables: Rule violations, event triggers, alerts.
Result: Consistent detection of known risks.
4.2 Unsupervised Anomaly Detection
Find the unusual before you know what to call it.
Identify patterns that differ from normal activity without requiring every anomaly to be predefined. Learn the baseline behavior of a scene and surface deviations that deserve attention.
Deliverables: Anomaly scores, deviation events, trend data.
Result: Earlier discovery of unknown risks.
4.3 Threat Escalation
Turn detected threats into action.
Automatically escalate high-priority events based on severity, confidence and defined response rules. Route alerts to the right people or systems so critical events do not remain buried in the feed.
Deliverables: Alerts, notifications, escalation logs.
Result: Faster response, clearer ownership, fewer missed threats.
4.4 Incident Reporting
Turn a detected incident into a record.
Generate structured incident reports from detected events, supporting evidence and relevant metadata. Bring the timeline, location, event details and associated footage together in one place.
Deliverables: Incident reports, summaries, audit trails.
Result: Consistent documentation, faster review, stronger accountability.
5.1 Heatmap Analysis
See where activity concentrates over time.
Visualize movement, presence and activity across defined areas to reveal where people, vehicles or events concentrate. Compare patterns across locations and time periods to understand how space is being used.
Deliverables: Heatmaps, dwell time, density maps.
Result: Clearer spatial patterns, better use of space.
5.2 Flow & Movement Analytics
Understand how things move, not just where they are.
Analyze movement across defined paths, zones and points to reveal how people or vehicles flow through an environment. Identify bottlenecks, preferred routes and changes in movement patterns.
Deliverables: Flow paths, counts, movement trends, KPIs
Result: Better movement planning, measurable flow performance.
5.3 Queue Management
Know when waiting becomes a problem.
Monitor queues, waiting times and service areas to identify where customer or operational flow slows down. Track queue length and duration to reveal when intervention is needed.
Deliverables: Queue counts, waiting times, alerts
Result: Shorter waits, smoother service, better customer experience.
5.5 Operational Intelligence
Measure operations through what your cameras already see.
Translate continuous video into operational metrics across people, processes, spaces and assets. Connect visual activity with measurable performance indicators to understand where operations are working—and where they aren't.
Deliverables: Dashboards, KPIs, utilization metrics, performance reports.
Result: Decisions grounded in what is actually happening on the ground.