Cutsio Blog

How to Find a Person Across Multiple Security Cameras

The fastest way to find a person across multiple security cameras is to export the relevant camera footage to Cutsio and search for their clothing description, build, and movement patterns across every angle simultaneously.

How do you find a person across multiple security cameras?

The fastest way to find a person across multiple security cameras is to export the relevant camera footage to Cutsio and search for their clothing description, build, and movement patterns across every angle simultaneously. Instead of following a suspect by scrubbing through each camera feed in sequence — entrance camera, hallway camera, aisle camera, exit camera — investigators can search once and get every appearance from every camera.

Finding a person across multiple cameras is one of the most time-consuming tasks in security investigations. A suspect enters through the main door, walks through the store, spends time in a specific aisle, and exits through a side door. Each segment is captured by a different camera. The traditional approach requires the investigator to spot the suspect on the entrance camera, note the timestamp, switch to the next camera, scrub to the approximate time, spot the suspect again, and repeat for every camera. For a typical retail path covering 5 cameras, this takes 15 to 30 minutes of focused scrubbing per suspect. Cutsio finds the person on every camera at once.

Why is following a person across cameras traditionally so difficult?

Person tracking across cameras is difficult because each camera captures a different angle, different lighting, and a different segment of the person's path. The investigator must identify the person on each camera based on visual characteristics — clothing color, build, gait — and mentally track their movement through the facility.

The time sync problem adds significant difficulty. Camera A records the person at 2:15:30. Camera B captures the same person at 2:16:45 if the cameras are perfectly synced, but real-world camera systems often have time drift. A camera that is 30 seconds off can cause the investigator to miss the person entirely when switching feeds.

The multi-camera monitoring problem is the biggest bottleneck. An investigator can only watch one feed at a time. To follow a person across 5 cameras, the investigator must switch between 5 feeds, noting timestamps on each and mentally reconstructing the path. For a 2-minute walk through a store, the reconstruction takes 15 to 30 minutes of manual work. For more on the search workflow, read our guide to searching security camera footage by description.

How does Cutsio find a person across multiple cameras?

Upload all relevant camera exports to Cutsio. Multimodal visual intelligence analyzes every frame of every camera feed, identifying people by their visual characteristics. Searching for a person's description returns every appearance across all cameras simultaneously.

| Search Query | What Cutsio Finds | Investigation Value |

|---|---|---|

| "person in blue jacket and jeans" | Every camera that captured that person | Full path reconstruction |

| "person with black backpack" | Bag-carrying individual across all feeds | Suspect identification |

| "tall person in red shirt" | Height and color combination | Distinctive appearance search |

| "person walking quickly toward exit" | Movement pattern matching | Flight behavior detection |

| "person near electronics display" | Location and appearance cross-reference | Area-specific search |

Searching for "person in blue jacket and jeans" across a Collection containing entrance, hallway, aisle, and exit cameras returns every match from every camera. The results are organized by camera and timestamp, automatically reconstructing the person's path through the facility. The investigator sees the full sequence — entrance at 2:15:30, hallway at 2:16:15, aisle at 2:17:00, exit at 2:17:45 — without scrubbing a single frame.

For suspects whose clothing is partially known, the investigator can combine multiple descriptors. "Person with black backpack and red shoes" narrows the search even when the jacket color is unknown. Each additional descriptor reduces false positives and speeds identification.

How do you track a person across a multi-floor facility?

Multi-floor facilities like department stores, office buildings, schools, and hospitals present an additional challenge. The person moves between floors using stairs or elevators, and each floor has its own set of cameras. Tracking them manually requires checking each floor's cameras in sequence.

Cutsio's cross-camera search handles multi-floor tracking the same as single-floor tracking. Export footage from every floor's cameras and search for the person's description in one query. The results show which floors the person visited, how long they stayed on each floor, and which route they took between floors.

For example, searching for "person in gray hoodie" across a 3-floor department store returns every appearance on every floor. The results reveal whether the person spent time on the floor where the theft occurred or moved between floors suspiciously. For more on multi-camera incident reconstruction, read our guide to creating an incident timeline from multi-camera footage.

How do you identify a person when clothing description is limited?

Not every investigation starts with a detailed clothing description. Sometimes the only information is "a man in a dark coat" or "someone wearing a hat." These limited descriptions still work with Cutsio's visual search, but the results will be broader.

For limited descriptions, the investigator combines the visual search with time and location filters. "Person in dark coat near electronics aisle between 2 PM and 3 PM" narrows the results to a specific area and time window. If 5 people match that description, the investigator can review the clips and identify the correct person by additional characteristics like height, build, or behavior.

For person-of-interest investigations where the investigator has a photo or sketch, Cutsio can search for visual similarity across the footage. Upload the reference image and search for visually similar people across all cameras.

How does per-minute pricing work for person search investigations?

Person search investigations typically involve short, targeted exports. Export 1 to 2 hours from 3 to 5 cameras — 3 to 10 hours total. At Cutsio's per-minute pricing, a single person search costs under $1.00 of indexed footage. The Pro plan at $59 per month covers dozens of person searches per month.

FAQ

How accurate is person search across cameras?

Accuracy depends on the distinctiveness of the person's appearance. A person wearing a bright red jacket is easier to track than someone in generic dark clothing. Combining multiple descriptors improves accuracy.

Can Cutsio track a person across cameras with different angles and lighting?

Yes. Visual intelligence matches people across different camera angles and lighting conditions by correlating visual characteristics like clothing color, build, and movement patterns.

How many cameras can I include in a single person search?

There is no limit. Include footage from any number of cameras in a Collection. Search across all of them simultaneously.

Can I track a person across different days?

Yes. If you have footage from multiple days uploaded, search for the person's description and Cutsio returns every appearance from every day.

Can I export the person's full path as a single video?

Yes. Compile the clips from each camera into a chronological timeline and export as a single MP4 file.