Head Reconstruction from Internet Photos
3D face reconstruction from Internet photos has recently produced exciting results. A person’s face, e.g., Tom Hanks, can be modeled and animated in 3D from a completely uncalibrated photo collection. Most methods, however, focus solely on face area and m
- PDF / 2,520,435 Bytes
- 15 Pages / 439.37 x 666.142 pts Page_size
- 24 Downloads / 200 Views
Abstract. 3D face reconstruction from Internet photos has recently produced exciting results. A person’s face, e.g., Tom Hanks, can be modeled and animated in 3D from a completely uncalibrated photo collection. Most methods, however, focus solely on face area and mask out the rest of the head. This paper proposes that head modeling from the Internet is a problem we can solve. We target reconstruction of the rough shape of the head. Our method is to gradually “grow” the head mesh starting from the frontal face and extending to the rest of views using photometric stereo constraints. We call our method boundary-value growing algorithm. Results on photos of celebrities downloaded from the Internet are presented. Keywords: Internet photo collections · Head modeling Unconstrained 3D reconstruction · Uncalibrated
1
· In the wild ·
Introduction “If two heads are better than one, then what about double chins? On that note, I will help myself to seconds.” —Jarod Kintz
Methods that reconstruct 3D models of people’s heads from images need to account for varying 3D pose, lighting, non-rigid changes due to expressions, relatively smooth surfaces of faces, ears and neck, and finally, the hair. Great reconstructions can be achieved nowadays in case the input photos are captured in a calibrated lab setting or semi-calibrated setup where the person has to participate in the capturing session (see related work). Reconstructing from Internet photos, however, is an open problem due to the high degree of variability across uncalibrated photos; lighting, pose, cameras and resolution change dramatically across photos. In recent years, reconstruction of faces from the Internet have received a lot of attention. All face-focused methods, however, mask out the head using a fixed face mask and focus only on the face area. For real-life applications, we must be able to reconstruct a full head. So what is it there to reconstruct except for the face? At the minimum, to create full head models we need to be able to reconstruct the ears, and at least part of the neck, chin, and overall head shape. Additionally, hair reconstruction is a difficult problem. One approach is to use morphable model methods. These, c Springer International Publishing AG 2016 B. Leibe et al. (Eds.): ECCV 2016, Part II, LNCS 9906, pp. 360–374, 2016. DOI: 10.1007/978-3-319-46475-6 23
Head Reconstruction from Internet Photos
361
however, do not fit the head explicitly but instead use fitting based on the face and provide a mostly average (non-personalized) bald model for the head. This paper addresses the new direction of head reconstruction directly from Internet data. We propose an algorithm to create a rough head shape, and frame the problem as follows. Given a photo collection, obtained by searching for photos of a specific person on Google image search, we would like to reconstruct a 3D model of that person’s head. Just like [1] (that focused only on the face area) we aim to reconstruct an average rigid model of the person from the whole collection. This model can be
Data Loading...
