Automated 3D Face Reconstruction from Multiple Images Using Quality Measures

Piotraschke, Marcel; Blanz, Volker

doi:10.1109/cvpr.2016.372

Cited by 80 publications

(62 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our confidence prediction subnet is also trained in a weakly-supervised fashion without ground-truth label. We show that our method clearly outperforms naive aggregation (e.g., shape averaging) and some heuristic strategies [34]. To our knowledge, this is the first attempt towards CNN-based 3D face reconstruction and aggregation from an unconstrained image set.…”

Section: Introductionmentioning

confidence: 79%

“…The whole pipeline is complex and may break down under severe occlusion and extreme pose. Our goal in this paper is not to replace these traditional methods, but to study the shape aggregation problem (similar to [34]) with a CNN and provide an extremely fast and robust alternative learned end-to-end.…”

Section: Related Workmentioning

confidence: 99%

“…To our knowledge, our method is the first one applying neutral networks for face reconstruction confidence prediction and aggregation. So here we compare with a heuristic strategy of Piotraschke and Blanz [34]. Table 6 shows that our method produced better results than shape averaging and [34] on the MICC dataset (we treat a sequence as an image set).…”

Section: Comparison With Prior Artmentioning

confidence: 99%

See 2 more Smart Citations

Accurate 3D Face Reconstruction With Weakly-Supervised Learning: From Single Image to Image Set

Deng

Yang

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)

644

583

View full text Add to dashboard Cite

Recently, deep learning based 3D face reconstruction methods have shown promising results in both quality and efficiency. However, training deep neural networks typically requires a large volume of data, whereas face images with ground-truth 3D face shapes are scarce. In this paper, we propose a novel deep 3D face reconstruction approach that 1) leverages a robust, hybrid loss function for weakly-supervised learning which takes into account both low-level and perception-level information for supervision, and 2) performs multi-image face reconstruction by exploiting complementary information from different images for shape aggregation. Our method is fast, accurate, and robust to occlusion and large pose. We provide comprehensive experiments on three datasets, systematically comparing our method with fifteen recent methods and demonstrating its state-of-the-art performance.

show abstract

Section: Introductionmentioning

confidence: 79%

Section: Related Workmentioning

confidence: 99%

Section: Comparison With Prior Artmentioning

confidence: 99%

See 1 more Smart Citation

Accurate 3D Face Reconstruction With Weakly-Supervised Learning: From Single Image to Image Set

Deng

Yang

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)

644

583

View full text Add to dashboard Cite

show abstract

“…The second one is to average the three predicted 3D models in a triplet and then compute the point-to-plane errors between the pooled 3D model with ground-truth model (shown in Table 2 as "+pool" entries). The third one is to compute the weighted average of three predicted 3D models as [22] Genova18 MoFA Tran17 Ours Inputs and then compute the point-to-plane errors (shown in Table 2 as "+ [22]" entries). Table 2 shows the mean errors of the comparison.…”

Section: Comparisons To State-of-the-art Methodsmentioning

confidence: 99%

“…They used RNNs to fuse identity-related features from CNNs to produce more discriminative reconstructions, but multi-view geometric constraints are not exploited in their approach. Notice that there are some other 3DMM-based methods in multi-image settings [22], but in these work each input image is dealt individually, which is not the same as our multi-view setting.…”

Section: Multi-view 3dmm-based Reconstructionmentioning

confidence: 99%

MVF-Net: Multi-View 3D Face Morphable Model Regression

Bao

Chen

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

122

View full text Add to dashboard Cite

We address the problem of recovering the 3D geometry of a human face from a set of facial images in multiple views. While recent studies have shown impressive progress in 3D Morphable Model (3DMM) based facial reconstruction, the settings are mostly restricted to a single view. There is an inherent drawback in the single-view setting: the lack of reliable 3D constraints can cause unresolvable ambiguities. We in this paper explore 3DMM-based shape recovery in a different setting, where a set of multi-view facial images are given as input. A novel approach is proposed to regress 3DMM parameters from multi-view inputs with an end-toend trainable Convolutional Neural Network (CNN). Multiview geometric constraints are incorporated into the network by establishing dense correspondences between different views leveraging a novel self-supervised view alignment loss. The main ingredient of the view alignment loss is a differentiable dense optical flow estimator that can backpropagate the alignment errors between an input view and a synthetic rendering from another input view, which is projected to the target view through the 3D shape to be inferred. Through minimizing the view alignment loss, better 3D shapes can be recovered such that the synthetic projections from one view to another can better align with the observed image. Extensive experiments demonstrate the superiority of the proposed method over other 3DMM methods.

show abstract

Analysis of 3D Facial Dysmorphology in Genetic Syndromes from Unconstrained 2D Photographs

Porras

Boyle

et al. 2018

Medical Image Computing and Computer Assisted Intervention – MICCAI 2018

View full text Add to dashboard Cite

Automated 3D Face Reconstruction from Multiple Images Using Quality Measures

Cited by 80 publications

References 29 publications

Accurate 3D Face Reconstruction With Weakly-Supervised Learning: From Single Image to Image Set

Accurate 3D Face Reconstruction With Weakly-Supervised Learning: From Single Image to Image Set

MVF-Net: Multi-View 3D Face Morphable Model Regression

Analysis of 3D Facial Dysmorphology in Genetic Syndromes from Unconstrained 2D Photographs

Contact Info

Product

Resources

About