Feature‐based probabilistic texture blending with feature variations for terrains

Abstract

The use of linear interpolation to blend different terrain types with distinct features produces translucency artefacts that can detract from the realism of the scene. The approach presented in this paper addresses the feature agnosticism of linear blending and makes the distinction between features (bricks, cobble stone, etc.) and non-features (cement, mortar, etc.). Using the blend weights from Bloom's texture splatting, intermittent texture transitions are generated on the fly without the need for artistic intervention. Furthermore, feature shapes are modified dynamically to give the illusion of wear and tear, thus further reducing repetition and adding authenticity to the scene. The memory footprint is constant regardless of texture complexity and uses nearly eight times less texture memory when compared to tile-based texture mapping. The scalability and diversity of our approach can be tailored to a wide range of hardware and can utilize textures of any size and shape compared to the grid layout and memory limitations of tile-based texture mapping.

Transition Contour Synthesis with Dynamic Patch Transitions

“…Whereas these details would be too ambiguous and indistinct for techniques such as FBPB [6], they contain enough contour information to exhibit the translucency artefacts demonstrated by Blend Maps tran- sitions. DPT masks these artefacts by creating the illusion of contour preservation to produce more sporadic, intermittent transition rather than the faded, uniform transitions of Blend Maps.…”

“…The quality and control of the blend however is limited by the resolution of the alpha blend masks, producing noticeable translucency artefacts (especially for salient material features). As such, efforts have been made to reduce these artifacts for salient information [8] or eliminate them entirely [6] by identifying salient details ('features') in realtime and blending them with full opacity in a stochastic manner. DPT instead increases the apparent resolution of the blend by introducing detailed variations along the contour of transitions, whilst at the same time keeping the performance and additional memory overhead consistent and competitive with conventional techniques.…”

“…This sampled blend value, common to all points within a patch, is interpreted as the probability of a given patch appearing (c). These patches are then drawn or discarded in a probabilistic manner [6] to produce a binary patch mask (d).…”

“…DPT builds on the previous work of Feature-Based Probabilistic Blending (FBPB) [6], expanding the concept of probabilistic blending to present a general purpose algorithm for material transitions with textures which contain non-salient details. The work presented here is an extension of a short paper recently published in CGI 2013 [7], with this paper detailing our approach, addressing minification aliasing and contrasting the performance and results of our algorithm to Blend Maps [8].…”

Detecting color boundaries on 3D surfaces by applying edge‐detection image filters on a quad‐remeshing

“…Users could control the appearance of the textures by drawing hints 19 or providing arrangement maps 20 . Some studies also explored methods for automatically blending textures across different materials 21 . These methods were further extended to the editing of geometric surfaces 22 .…”