Multi-Camera Tile Mapping for Cross-View Object Handover

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Multi-camera systems face challenges in efficiently tracking and detecting objects across non-overlapping camera fields of view due to lack of synchronization and camera movements, leading to inaccuracies in object detection and handover of streaming feeds.

Innovation Solution

A multi-camera system divides camera coverage into a set of tiles, uses a view mapping database to identify corresponding tiles across cameras, and shares this information to enable seamless object tracking and streaming handover.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If multi-camera systems track objects across non-overlapping fields of view, then object tracking accuracy improves, but computational load and complexity increase

Engineering Contradiction:
Improveobject tracking accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent divides the camera coverage area into a grid of tiles, where each tile represents a specific spatial region. By segmenting the field of view into discrete tiles, the system can efficiently track objects across non-overlapping camera views by comparing tile coordinates rather than processing entire images. This segmentation approach reduces computational complexity while maintaining tracking accuracy across multiple cameras.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If feature extraction is performed across multiple cameras for object detection, then detection accuracy improves, but processing time and computational resources increase

Engineering Contradiction:
Improvedetection accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent extracts only the necessary information for tracking - specifically tile coordinates and object locations within tiles - rather than performing extensive feature extraction on entire images. By taking out only the essential spatial information needed for cross-camera tracking, the system achieves accurate object detection while significantly reducing processing time and computational resource requirements.

Inventive Principle:
Principle #2Taking out (Extraction)

3Reliability

If camera coverage is divided into tiles for mapping, then object handover between cameras improves, but data processing complexity increases

Engineering Contradiction:
Improvestreaming handover reliabilityVSAvoidmapping data complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent applies local quality by creating simplified tile mappings for specific regions of interest rather than processing entire camera fields of view. Each tile contains localized information about object presence and coordinates, allowing for reliable streaming handover decisions to be made based on local tile data rather than global image analysis. This approach improves handover reliability while keeping the mapping data manageable.

Inventive Principle:
Principle #3Local quality

Data Source

PatentUS20260025483A1Multi-camera image data processing
Publication Date: 2026.01.22 NOKIA TECHNOLOGIES OY
  • US20260025483A1 patent drawing
  • US20260025483A1 patent drawing
  • US20260025483A1 patent drawing

AI summary

There is provided method for a multi-camera system comprising at least a first and a second camera, wherein coverage of the first and the second camera is divided into a set of tiles. The method comprises detecting an object in a first region on the first camera's field of view, FOV, and determining a location of the object in terms of one or more tiles. The method further comprises requesting a mapping information identifying one or more tiles in the second camera's FOV that correspond to the one or more tiles in the first camera's FOV, and receiving a response from a view mapping database that is configured to identify the one or more tiles in the second camera's FOV that correspond to the one or more tiles in the first camera's FOV, where the response further identifies one or more additional tiles contiguous with the identified one or more tiles in the second camera's FOV. The response is shared from the first camera to the second camera such that the second camera is enabled to find the object in the one or more of the tiles identified in the response.