3D Image Encoding Using Variable Occlusion Layer Geometry

The patent applies dimensionality change by allowing occlusion layers to have different sizes and shapes compared to the foreground layer. Specifically, occlusion layers can be larger in area to capture hidden background information that would otherwise be lost. The scaling data parameter enables the system to handle these dimensional variations by providing transformation information to map between different layer geometries, thus resolving the contradiction between representation accuracy and structural simplicity.

The patent employs parameter changes by introducing scaling data as a key parameter that describes the geometric relationship between foreground and occlusion layers. This parameter enables the system to handle variable layer sizes and shapes systematically. By encoding the scaling relationship rather than processing arbitrary geometric transformations, the patent reduces data processing complexity while maintaining the ability to represent occluded information accurately in enlarged occlusion layers.

The patent introduces scaling data as an intermediary element that mediates between multiple layers with different viewing areas. This intermediary parameter provides a systematic way to describe and process the geometric relationships between layers during encoding and decoding. Rather than implementing complex geometric transformations directly, the system uses the scaling data intermediary to manage information flow between layers, thus capturing complete scene information while reducing processing complexity.