Browsing by Author "Lai, Yukun"

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    Generating 3D Faces using Multi-column Graph Convolutional Networks
    (The Eurographics Association and John Wiley & Sons Ltd., 2019) Li, Kun; Liu, Jingying; Lai, Yu-Kun; Yang, Jingyu; Lee, Jehee and Theobalt, Christian and Wetzstein, Gordon
    In this work, we introduce multi-column graph convolutional networks (MGCNs), a deep generative model for 3D mesh surfaces that effectively learns a non-linear facial representation. We perform spectral decomposition of meshes and apply convolutions directly in the frequency domain. Our network architecture involves multiple columns of graph convolutional networks (GCNs), namely large GCN (L-GCN), medium GCN (M-GCN) and small GCN (S-GCN), with different filter sizes to extract features at different scales. L-GCN is more useful to extract large-scale features, whereas S-GCN is effective for extracting subtle and fine-grained features, and M-GCN captures information in between. Therefore, to obtain a high-quality representation, we propose a selective fusion method that adaptively integrates these three kinds of information. Spatially non-local relationships are also exploited through a self-attention mechanism to further improve the representation ability in the latent vector space. Through extensive experiments, we demonstrate the superiority of our end-to-end framework in improving the accuracy of 3D face reconstruction. Moreover, with the help of variational inference, our model has excellent generating ability.
  • Thumbnail Image
    Item
    An Image-based Approach for Detecting Faces Carved in Heritage Monuments
    (The Eurographics Association, 2018) Lai, Yu-Kun; Echavarria, Karina Rodriguez; Song, Ran; Rosin, Paul L.; Sablatnig, Robert and Wimmer, Michael
    Heritage monuments such as columns, memorials and buildings are typically carved with a variety of visual features, including figural content, illustrating scenes from battles or historical narratives. Understanding such visual features is of interest to heritage professionals as it can facilitate the study of such monuments and their conservation. However, this visual analysis can be challenging due to the large-scale size, the amount of carvings and difficulty of access to monuments across the world. This paper makes a contribution towards this goal by presenting work-in-progress for developing image-based approaches for detecting visual features in 3D models, in particular of human faces. The motivation for focusing on faces is the prominence of human figures throughout monuments in the world. The methods are tested on a 3D model of a section of the Trajan Column cast at the Victoria and Albert (V&A) Museum in London, UK. The initial results suggest that methods based on machine learning can provide useful tools for heritage professionals to deal with the large-scale challenges presented by such large monuments.
  • Thumbnail Image
    Item
    SHREC 2020 Track: Non-rigid Shape Correspondence of Physically-Based Deformations
    (The Eurographics Association, 2020) Dyke, Roberto M.; Zhou, Feng; Lai, Yu-Kun; Rosin, Paul L.; Guo, Daoliang; Li, Kun; Marin, Riccardo; Yang, Jingyu; Schreck, Tobias and Theoharis, Theoharis and Pratikakis, Ioannis and Spagnuolo, Michela and Veltkamp, Remco C.
    Commonly, novel non-rigid shape correspondence techniques focus on particular matching challenges. This can lead to the potential trade-off of poorer performance in other scenarios. An ideal dataset would provide a granular means for degrees of evaluation. In this paper, we propose a novel dataset of real scans that contain challenging non-isometric deformations to evaluate non-rigid point-to-point correspondence and registration algorithms. The deformations included in our dataset cover extreme types of physically-based contortions of a toy rabbit. Furthermore, shape pairs contain incrementally different types and amounts of deformation, this enables performance to be systematically evaluated with respect to the nature of the deformation. A brief investigation into different methods for initialising correspondence was undertaken, and a series of experiments were subsequently conducted to investigate the performance of state-of-the-art methods on the proposed dataset. We find that methods that rely on initial correspondences and local descriptors that are sensitive to local surface changes perform poorly in comparison to other strategies, and that a template-based approach performs the best.
  • Thumbnail Image
    Item
    Simultaneous Multi-Attribute Image-to-Image Translation Using Parallel Latent Transform Networks
    (The Eurographics Association and John Wiley & Sons Ltd., 2020) Xu, Sen-Zhe; Lai, Yu-Kun; Eisemann, Elmar and Jacobson, Alec and Zhang, Fang-Lue
    Image-to-image translation has been widely studied. Since real-world images can often be described by multiple attributes, it is useful to manipulate them at the same time. However, most methods focus on transforming between two domains, and when they chain multiple single attribute transform networks together, the results are affected by the order of chaining, and the performance drops with the out-of-domain issue for intermediate results. Existing multi-domain transfer methods mostly manipulate multiple attributes by adding a list of attribute labels to the network feature, but they also suffer from interference of different attributes, and perform worse when multiple attributes are manipulated. We propose a novel approach to multiattribute image-to-image translation using several parallel latent transform networks, where multiple attributes are manipulated in parallel and simultaneously, which eliminates both issues. To avoid the interference of different attributes, we introduce a novel soft independence constraint for the changes caused by different attributes. Extensive experiments show that our method outperforms state-of-the-art methods.