Entity

Time filter

Source Type


Xie X.,CAS Shenzhen Institutes of Advanced Technology | Xie X.,Shenzhen VisuCA Key Laboratory
Signal Processing | Year: 2014

The Empirical Mode Decomposition (EMD) can adaptively decompose a complex signal into Intrinsic Mode Functions (IMFs) that are relevant to intrinsic physical significances, therefore is a powerful tool for multi-scale analysis of non-stationary signals. Towards restoring a frontal-illuminated face from a single image, in this paper we study the usage of EMD for manipulating the illumination issue on face images. We propose an EMD-based algorithm to extract the illumination-insensitive facial features. We also come up with an EMD-based scheme to detect the shadows and to reduce the effects of shadows on face images. By preserving the intrinsic facial features as well as lessening the shadows, it is more likely to restore the frontal-illuminated face image with good visual quality from a single image. Experiments verify the effectiveness of the proposed methods. © 2013 Elsevier B.V. Source


Yuan Q.,South China University of Technology | Li G.,South China University of Technology | Xu K.,Shenzhen VisuCA Key Laboratory | Xu K.,National University of Defense Technology | And 3 more authors.
Computer Graphics Forum | Year: 2016

Consistent segmentation is to the center of many applications based on dynamic geometric data. Directly segmenting a raw 3D point cloud sequence is a challenging task due to the low data quality and large inter-frame variation across the whole sequence. We propose a local-to-global approach to co-segment point cloud sequences of articulated objects into near-rigid moving parts. Our method starts from a per-frame point clustering, derived from a robust voting-based trajectory analysis. The local segments are then progressively propagated to the neighboring frames with a cut propagation operation, and further merged through all frames using a novel space-time segment grouping technqiue, leading to a globally consistent and compact segmentation of the entire articulated point cloud sequence. Such progressive propagating and merging, in both space and time dimensions, makes our co-segmentation algorithm especially robust in handling noise, occlusions and pose/view variations that are usually associated with raw scan data. © 2016 The Author(s) Computer Graphics Forum © 2016 The Eurographics Association and John Wiley & Sons Ltd. Published by John Wiley & Sons Ltd. Source


Xie X.,CAS Shenzhen Institutes of Advanced Technology | Xie X.,Shenzhen VisuCA Key Laboratory | Xu K.,CAS Shenzhen Institutes of Advanced Technology | Xu K.,National University of Defense Technology | And 9 more authors.
Computer Graphics Forum | Year: 2013

Designing 3D objects from scratch is difficult, especially when the user intent is fuzzy and lacks a clear target form. We facilitate design by providing reference and inspiration from existing model contexts. We rethink model design as navigating through different possible combinations of part assemblies based on a large collection of pre-segmented 3D models. We propose an interactive sketch-to-design system, where the user sketches prominent features of parts to combine. The sketched strokes are analysed individually, and more importantly, in context with the other parts to generate relevant shape suggestions via adesign galleryinterface. As a modelling session progresses and more parts get selected, contextual cues become increasingly dominant, and the model quickly converges to a final form. As a key enabler, we use pre-learned part-based contextual information to allow the user to quickly explore different combinations of parts. Our experiments demonstrate the effectiveness of our approach for efficiently designing new variations from existing shape collections. Designing 3D objects from scratch is difficult, especially when the user intent is fuzzy and lacks a clear target form. We facilitate design by providing reference and inspiration from existing model contexts. We rethink model design as navigating through different possible combinations of part assemblies based on a large collection of pre-segmented 3D models. We propose an interactive sketch-to-design system, where the user sketches prominent features of parts to combine. The sketched strokes are analyzed individually, and more importantly, in context with the other parts to generate relevant shape suggestions via a design gallery interface. As a modeling session progresses and more parts get selected, contextual cues become increasingly dominant, and the model quickly converges to a final form. As a key enabler, we use pre-learned part-based contextual information to allow the user to quickly explore different combinations of parts. © 2013 The Authors Computer Graphics Forum © 2013 The Eurographics Association and John Wiley & Sons Ltd. Source


Yin K.,Shenzhen VisuCA Key Laboratory | Huang H.,Shenzhen VisuCA Key Laboratory | Long P.,Shenzhen VisuCA Key Laboratory | Gaissinski A.,Tel Aviv University | And 2 more authors.
Computer Graphics Forum | Year: 2016

Digitally capturing vegetation using off-the-shelf scanners is a challenging problem. Plants typically exhibit large self-occlusions and thin structures which cannot be properly scanned. Furthermore, plants are essentially dynamic, deforming over the time, which yield additional difficulties in the scanning process. In this paper, we present a novel technique for acquiring and modelling of plants and foliage. At the core of our method is an intrusive acquisition approach, which disassembles the plant into disjoint parts that can be accurately scanned and reconstructed offline. We use the reconstructed part meshes as 3D proxies for the reconstruction of the complete plant and devise a global-to-local non-rigid registration technique that preserves specific plant characteristics. Our method is tested on plants of various styles, appearances and characteristics. Results show successful reconstructions with high accuracy with respect to the acquired data. Digitally capturing vegetation using off-the-shelf scanners is a challenging problem. Plants typically exhibit large self-occlusions and thin structures which cannot be properly scanned. Furthermore, plants are essentially dynamic, deforming over the time, which yield additional difficulties in the scanning process. In this paper, we present a novel technique for acquiring and modelling of plants and foliage. At the core of our method is an intrusive acquisition approach, which disassembles the plant into disjoint parts that can be accurately scanned and reconstructed offline. © 2015 The Eurographics Association and John Wiley & Sons Ltd. Source


Xie Z.,Shenzhen VisuCA Key Laboratory | Xie Z.,National University of Defense Technology | Xu K.,Shenzhen VisuCA Key Laboratory | Xu K.,National University of Defense Technology | And 4 more authors.
Computer Graphics Forum | Year: 2015

Feature learning for 3D shapes is challenging due to the lack of natural paramterization for 3D surface models. We adopt the multi-view depth image representation and propose Multi-View Deep Extreme Learning Machine (MVD-ELM) to achieve fast and quality projective feature learning for 3D shapes. In contrast to existing multi-view learning approaches, our method ensures the feature maps learned for different views are mutually dependent via shared weights and in each layer, their unprojections together form a valid 3D reconstruction of the input 3D shape through using normalized convolution kernels. These lead to a more accurate 3D feature learning as shown by the encouraging results in several applications. Moreover, the 3D reconstruction property enables clear visualization of the learned features, which further demonstrates the meaningfulness of our feature learning. © 2015 The Author(s) Computer Graphics Forum. Source

Discover hidden collaborations