View Stabilisation

Deep Vision's novel concept of data abstraction quickly transforms abundant sensor data into a form that is easily classified and efficiently analysed.

The abstractions created from the raw data accurately describe the content of the data. By utilising the descriptive properties of the abstractions, coupled with their position on the sensor and relative positions over time, content-based motion vectors are formed.

Content-based motion vectors allow for more accurately global and partial image alignment than those of pixel-based motion vectors. The content-based motion vectors can then be used to remove high and low frequency motion from the view.

Additionally, Deep Vision's data abstraction technology operates with a throughput of 100+ frames per second.

With prior knowledge of the target objects, or their characteristics, Deep Vision's view stabilising technology delivers a highly robust, real-time target locking and stabilising solution.

• Operator Assistance
• Target locking (e.g. Centre target on screen)
• Enhanced video encoding compression

Features

• Content-based view analysis
• Environment independent
• No calibration required

• Operating System: Any (GNU/Linux recommended)
• Hardware Requirements: None
• Sensor Modalities: Visual, Thermal, Sonar
• Timings†: 100+ FPS
• Runtime Memory Requirements†: 300 KB
• Storage Requirements‡: 1.1 KB

† Typical. Based on a 640 x 480 data set
‡ Typical. Based on 45 abstractions (avg. 25 symbols each)

Input Requirements

• Archived videos and images
• Real-time acquisition from visual, thermal, or sonar sensors.