Single-view hair modeling for portrait manipulation

Abstract

Figure 1: Given a portrait image and a few strokes drawn by the user as input (a), our method generates a strand-based 3D hair model as shown in (b), where a fraction of reconstructed fibers are highlighted. The hair model can be used to convert the input portrait into a pop-up model (c) which can be rendered in a novel view (d). It also enables several interesting applications such as transferring the hairstyle of one subject to another (e). Original images courtesy of Getty Images (a) and Andrew MacPherson (e). AbstractHuman hair is known to be very difficult to model or reconstruct. In this paper, we focus on applications related to portrait manipulation and take an application-driven approach to hair modeling. To enable an average user to achieve interesting portrait manipulation results, we develop a single-view hair modeling technique with modest user interaction to meet the unique requirements set by portrait manipulation. Our method relies on heuristics to generate a plausible high-resolution strand-based 3D hair model. This is made possible by an effective high-precision 2D strand tracing algorithm, which explicitly models uncertainty and local layering during tracing. The depth of the traced strands is solved through an optimization, which simultaneously considers depth constraints, layering constraints as well as regularization terms. Our single-view hair modeling enables a number of interesting applications that were previously challenging, including transferring the hairstyle of one subject to another in a potentially different pose, rendering the original portrait in a novel view and image-space hair editing.

Literature Review of Facial Skin Embellishment Using Flexible Region Aware Mask

“…Mostly the users requirement is immediately do facial beautification with the minimum number of operations to avoid endless manipulations. There for it is useful to develop a face image enhancement technique that is effective ,convenient, and flexible.The increasing demand for face image retouching has worked on many methods,such as facial geometric beautification [34],digital facial makeup [29], personal photo enhancement [20], and hair modeling [21].…”

Wide-Baseline Hair Capture Using Strand-Based Refinement

“…Guided by hair simulation, Zhang et al [29] reconstructed smooth hair dynamics with 7 cameras. Using only a single view, Chai et al [4] proposed a method to generate a depth map for convincing view interpolation of different hairstyles. Beeler at al.…”

“…The orientation map is enhanced with 3 passes of iterative refinement to improve the signal-to-noise ratio as in [4]. To further reduce noises in regions with low confidence, we apply the bilateral filtering method in [20] to diffuse the orientations of the high confidence regions.…”

“…4(b)) using hysteresis thresholding similar to [4]. The result is a set of poly-line 2D strands consisting of densely sampled vertices in about 1-pixel steps.…”

“…The formulation for strand level smoothness E strand stems from the discrete elastic rod model [3] often used in hair simulation that minimizes the squared curvature along hair strands. Further inspired by [4], we take into account the orientation similarity in the bilateral same-view weight w + so that the Figure 8: The illustration of silhouette energy. The sign of (p h(p)) · n h determines if a point is inside H and thus the value of b in silhouette energy for p in and p out .…”