Single-picture reconstruction and rendering of trees for plausible vegetation synthesis

Abstract

State-of-the-art approaches for tree reconstruction either put limiting constraints on the input side (requiring multiple photographs, a scanned point cloud or intensive user input) or provide a representation only suitable for front views of the tree. In this paper we present a complete pipeline for synthesizing and rendering detailed trees from a single photograph with minimal user effort. Since the overall shape and appearance of each tree is recovered from a single photograph of the tree crown, artists can benefit from georeferenced images to populate landscapes with native tree species. A key element of our approach is a compact representation of dense tree crowns through a radial distance map. Our first contribution is an automatic algorithm for generating such representations from a single exemplar image of a tree. We create a rough estimate of the crown shape by solving a thin-plate energy minimization problem, and then add detail through a simplified shape-from-shading approach. The use of seamless texture synthesis results in an image-based representation that can be rendered from arbitrary view directions at different levels of detail. Distant trees benefit from an output-sensitive algorithm inspired on relief mapping. For close-up trees we use a billboard cloud where leaflets are distributed inside the crown shape through a space colonization algorithm. In both cases our representation ensures efficient preservation of the crown shape. Major benefits of our approach include: it recovers the overall shape from a single tree image, involves no tree modeling knowledge and minimal authoring effort, and the associated image-based representation is easy to compress and thus suitable for network streaming.Peer ReviewedPostprint (author's final draft

Image‐Based Tree Variations

“…Then, we position these billboards to populate the tree crown. Since we want the front view to be as close as possible to the original photograph, we will simply respect their relative positions on the image plane, and compute an extruded depth based on a bi‐harmonic equation with the silhouette constrained to lie on the image plane [ACA16]. Since this equation results in a smooth surface, we improve the positions by deepening the billboards with mean luminance smaller than the overall mean luminance.…”

“…For dense trees, it is possible to extract more compact representations. Argudo et al [ACA16] present a complete pipeline for synthesizing and rendering detailed trees. A rough estimate of the crown shape is created by solving a thin-plate energy minimization problem.…”

“…Vegetation (trees, bushes and grass) is an essential part of natural outdoor scenes. Many different plant representations have been proposed in the literature, including polygonal‐based, image‐based [BCF*05], point‐based [DCSD02], volume‐based [DN04], relief‐mapping‐based [ACA16] and procedural [ACV*14] approaches. Here, we focus exclusively on image‐based representations of trees.…”

Interactive Large‐Scale Procedural Forest Construction and Visualization Based on Particle Flow Simulation

“…The furthest plants, such as grass [PC01] or trees [BZCJ09], can be modelled with billboards, nearer ones with simplified polygons [DGMW11] and the nearest at full resolution. Alternatively, distant trees can be visualized using relief mapping and the nearest ones using billboard clouds [ACA16].…”

Algorithm development for pattern generation inspired from tree growth

“…the perimeter and an imprecise topography of tree which was fed to the SCA as the initial 3D target area and the SCA developed the tree shape (Argudo, Chica, & Andujar, 2016). Figure 10 shows the application of algorithm in generating tree shapes which is later used in image analysis.…”