Guiqing Li

South China University of Technology | SCUT · Department of Computer Science and Technology

Recently, single-image SVBRDF capture is formulated as a regression problem, which uses a network to infer four SVBRDF maps from a flash-lit image. However, the accuracy is still not satisfactory since previous approaches usually adopt endto-end inference strategies. To mitigate the challenge, we propose “auxiliary renderings” as the intermediate r...

In this study, we present a new and innovative framework for acquiring high-quality SVBRDF maps. Our approach addresses the limitations of the current methods and proposes a new solution. The core of our method is a simple hardware setup consisting of a consumer-level camera, LED lights, and a carefully designed network that can accurately obtain t...

Although there has been some progress in 3D human pose and shape estimation, accurately predicting complex human poses is still challenging. To tackle this issue and improve the accuracy of the human mesh reconstruction, we propose an end-to-end framework called Multi-level Attention Network (MANet). MANet consists of three modules: Intra Part Atte...

It is well known that regressing the parametric representation of a human body from a single image suffers low accuracy due to sparse information use and error accumulation. Although being able to achieve higher accuracy by avoiding these issues, directly regressing vertices may result in vertex outliers. We present METRO-X, a novel method for reco...

Previous methods for human motion generation have predominantly relied on skeleton representations to depict human poses and motion. These methods typically use a series of skeletons to represent the motion of a human. However, they are not directly suitable for handling the 3D point cloud sequences obtained from optical motion capture. To address...

We propose a GAN-based facial expression transfer method. It can transfer the facial expression of a reference subject to the source subject while preserving the source identity attributes, such as shape, appearance, and illumination. Our method consists of two modules based on GAN: Parts Generation Networks (PGNs), and Parts Fusion Network (PFN)....

Constructing a convex hull for the pixel colors of an image by viewing them as 3D points can extract a set of palette colors for the image, then image recoloring can be achieved by modifying the palette colors. For better recoloring effect, the convex hull should contain more pixels (inclusive) and be more compact. Otherwise, reconstruction error w...

Color pencil drawing is well-loved due to its rich expressiveness. This paper proposes an approach for generating feature-preserving color pencil drawings from photographs. To mimic the tonal style of color pencil drawings, which are much lighter and have relatively lower saturation than photographs, we devise a lightness enhancement mapping and a...

Human non‐speech sounds occur during expressions in a real‐life environment. Realising a person's incapability to prompt confident expressions by non‐speech sounds may assist in identifying premature disorder in medical applications. A novel dataset named Nonspeech7k is introduced that contains a diverse set of human non‐speech sounds, such as the...

For 3D mesh pose transfer, the target model is obtained by transferring the pose of the reference mesh to the source mesh, where the shape and pose of the source are usually different from that of the reference. In this paper, pose transfer is considered as a deformation process of the source mesh, and we propose a 3D mesh pose transfer method base...

Recently, the deep learning-based object detection methods have achieved a great success. However, the performance of such techniques deteriorates on the images captured under adverse weather conditions. To tackle this problem, a detection-driven enhancement network (DENet) which consists of three key modules for object detection is proposed. By us...

Simulating liquid-textile interaction has received great attention in computer graphics recently. Most existing methods take textiles as particles or parameterized meshes. Although these methods can generate visually pleasing results, they cannot simulate water content at a microscopic level due to the lack of geometrically modeling of textile's an...

Diverse human motion prediction aims at predicting multiple possible future pose sequences from a sequence of observed poses. Previous approaches usually employ deep generative networks to model the conditional distribution of data, and then randomly sample outcomes from the distribution. While different results can be obtained, they are usually th...

This paper presents a high-quality human motion prediction method that accurately predicts future human poses given observed ones. Our method is based on the observation that a good initial guess of the future poses is very helpful in improving the forecasting accuracy. This motivates us to propose a novel two-stage prediction framework, including...

While self-attention has been successfully applied in a variety of natural language processing and computer vision tasks, its application in Monte Carlo (MC) image denoising has not yet been well explored. This paper presents a self-attention based MC denoising deep learning network based on the fact that self-attention is essentially non-local mea...

Garment generation or reconstruction is becoming extremely demanding for many digital applications, and the traditional process is time-consuming. In recent years, garment reconstruction from sketch leveraging deep learning and principal component analysis (PCA) has made great progress. In this paper, we present a data-driven approach wherein 3D ga...

Camera focal length estimation from a single image is of great importance for many computer vision tasks. Unfortunately, previous methods cannot achieve satisfactory accuracy yet. This paper proposes a focal length estimation approach based on distances among four points in the scene of a single image. The issue is cast into a nonlinear optimizatio...

This paper investigates the estimate of motion parameters from 3D hand joint positions. We formulate the issue as an inverse kinematics problem with biomechanical constraints and propose a fast and robust iterative approach to address the constrained optimization. It elaborately designs a coordinate descent algorithm to decompose the problem into a...

Human motion prediction is a challenging task due to the stochasticity and aperiodicity of future poses. Recently, graph convolutional network has been proven to be very effective to learn dynamic relations among pose joints, which is helpful for pose prediction. On the other hand, one can abstract a human pose recursively to obtain a set of poses...

In this paper, we propose $\text{HF}^2$-VAD, a Hybrid framework that integrates Flow reconstruction and Frame prediction seamlessly to handle Video Anomaly Detection. Firstly, we design the network of ML-MemAE-SC (Multi-Level Memory modules in an Autoencoder with Skip Connections) to memorize normal patterns for optical flow reconstruction so that...

Volume segmentation based on surface attributes is an essential problem in multi-material fabrication and model packing. In practice, current mainstream fabrication techniques have difficulties in yielding models with diverse surface attributes in one pass owing to their craft limitations, making model segmentation a sensible choice for model reali...

Parameterizing Variations of human shapes and motions is a long-standing problem in computer graphics and vision. Most of the existing methods only deal with a specific kind of motion, such as body poses, facial expressions, or hand gestures. We propose PanoMan (sParse locAlized compoNents based mOdel for full huMAn motioNs) to handle shape variati...

Human motion transfer has many applications in human behavior analysis, training data augmentation, and personalization in mixed reality. We propose a Body-Parts-Aware Generative Adversarial Network (BPA-GAN) for image-based human motion transfer. Our key idea is to take advantage of the human body with segmented parts instead of using the human sk...

Pairwise orthogonal planes are important semantic structures in indoor scenes. The method of Manhattan world orientation calculation and constrained planes selection can solve most segmentation problems of point cloud, but it is inefficient when it comes to small areas on beams, columns or other small architectural primitives. In this paper, we pro...

Determining optimal target tooth arrangements is a key step of treatment planning in digital orthodontics. Existing practice for specifying the target tooth arrangement involves tedious manual operations with the outcome quality depending heavily on the experience of individual specialists, leading to inefficiency and undesirable variations in trea...

Commodity-level scan cameras generally capture RGB-D image with depth missing or incorrect depth points if the surface of the object is transparent, bright, or black. These incorrect depth points are generated randomly and limit the downstream applications of raw RGB-D images. In this paper, we propose a coarse-to-fine method to detect and eliminat...

Pose transfer between two 2D cartoon characters provides a fast way to copy pose without complex deformation operations on the 2D shape. This paper proposes an effective method for transferring the pose of 2D human cartoon characters while preserving the character’s geometric features. We compare our method with other similar works and discuss the...

This paper presents a self-supervised learning network called SP-Flow to generate keypoints in real-time for SLAM systems. Optical flows are employed to match the keypoints between two successive frames in the training process of SP-Flow. This approach enables the network to use datasets without manual annotations. To show the efficacy of our SP-Fl...

Hair modeling plays an important role in computer animation, virtual reality, and other applications. This paper proposes an encoder‐decoder network, named HAO‐CNN, to recover 3D hair strand models from a single image. Specifically, HAO‐CNN generates a volumetric vector field (VVF) from the oriented map of hairstyles. However, instead of directly w...

In digital furniture design, skillful designers usually use professional software to create new furniture designs with various textures and then take advantage of rendering tools to produce eye-catching design results. Generally, a fine- grained furniture model holds many geometric details, inducing significant time cost to model rendering and larg...

We present a method for synopsizing multiple videos captured by a set of surveillance cameras with some overlapped field-of-views. Currently, object-based approaches that directly shift objects along the time axis are already able to compute compact synopsis results for multiple surveillance videos. The challenge is how to present the multiple syno...

Video stabilization is usually composed of three stages: feature trajectory extraction, trajectory smoothing, and frame warping. Most previous approaches view them as three separate stages. This paper proposes a method combining the last two stages, namely the trajectory smoothing and frame warping stages, into a single optimization framework. The...

This paper proposes a framework for reconstructing 3D human head models from a single image. Firstly, a preprocessing system for images is designed. The image is automatically segmented using convolutional neural network. The Gabor filter is used to extract the direction of the hair, and a optional manual interaction is used to modify the results....

We propose an interactive contour extraction method inspired by a skill often adopted in sketching: an artist usually sketches an object by firstly drawing lots of short, directional, and redundant strokes, then following these small strokes to draw the final outline of the object. Our method simulates this process. To extract a contour, our method...

Sparse localized components are powerful tool for compactly representing time-varying geometries. This paper proposes an adaptive, sparse and localized decomposition for the animated meshes of a motion object encoded by edge lengths of triangles and dihedral angles between adjacent triangles. During the sparse localized decomposition, it employs Ga...

In [1] , the biographies of Hanqiu Sun and Guiqing Li included incorrect information. The correct biographies are as follows.

3D geometric features represent rich details of 3D models, whose scale is much larger than noise. Traditional point cloud denoising methods cannot handle the task of processing and analyzing these features. Rolling guidance normal filtering is proved to be a useful tool in image and mesh small features removing. However, its direct extension to poi...

In this paper, we propose to integrate multi-video object synopsis with the display method of optimal multi-view switching. We group occurrences of the same object in different cameras together and operate them in the same way, based on which we propose a unified framework that optimizes object-based synopsis variables and multi-view switching vari...

This paper presents a new two-step color transfer method which includes color mapping and detail preservation. To map source colors to target colors, which are from an image or palette, the proposed similarity-preserving color mapping algorithm uses the similarities between pixel color and dominant colors as existing algorithms and emphasizes the s...

Feature-preserving filtering of noisy point clouds plays a fundamental role in geometric processing. Though the guided filter is known to be a powerful tool for edge-aware image processing and mesh denoising, extending it to point clouds is not a trivial task due to the difficulty of defining a piecewise smooth normal field on point clouds with sha...

This paper proposes a new easy and fast 3D avatar reconstruction method using an RGB-D sensor. Users can easily implement human body scanning and modeling just with a personal computer and a single RGB-D sensor such as a Microsoft Kinect within a small workspace in their home or office. To make the reconstruction of 3D avatars easy and fast, a new...

Current face alignment algorithms can robustly find a set of landmarks along face contour. However, the landmarks are sparse and lack curve details, especially in chin and cheek areas where a lot of concave-convex bending information exists. In this paper, we propose a local to global seam cutting and integrating algorithm (L2GSCI) to extract conti...

To improve the reconstruction quality of vectorization images, an error-controllable vectorization algorithm based on subdivision surfaces is proposed. Given an image to be vectorized, it first extracts feature lines, constructs a dense initial mesh, and simplifies the initial mesh using the quadric error metric algorithm to obtain a base mesh. A L...

Stitching videos captured by handheld devices is very useful, but also very challenging due to the heavy and independent shakiness in the videos. In this paper, we propose a hand-taken video stitching method which combines the techniques of video stitching and stabilization together into a unified optimization framework. In this way, our method can...

Compactly representing time-varying geometries is an important issue in dynamic geometry processing. This paper proposes a framework of sparse localized decomposition for given animated meshes by analyzing the variation of edge lengths and dihedral angles (LAs) of the meshes. It first computes the length and dihedral angle of each edge for poses an...

Wide-baseline street image interpolation is useful but very challenging. Existing approaches either rely on heavyweight 3D reconstruction or computationally intensive deep networks. We present a lightweight and efficient method which uses simple homography computing and refining operators to estimate piecewise smooth homographies between input view...

In this paper, we propose a new fast normal-based interpolating subdivision scheme for curve and surface design. Different from the 4-points interpolating subdivision scheme, it is based on cubic Bezier curves and the normal vectors are used to generate a circle. Both a convex edge and an inflexion edge can be subdivided into convex sub-edges and t...

Colour plays an important role in interior design. A good colour scheme can usually convey some particular design philosophy and define a specific household style. Given the planar shape and functional type of a room, and a set of furniture for which a subset has a specified colour, this paper proposes a data-driven approach to automatically assign...

Rig-space physics a finite element method(FEM) based simulation technique that aims at adding secondary motion on a character while maintaining seamless cooperation with traditional animation pipelines. We enhance the rig-space physics by introducing several techniques, including general field interaction, proportional-derivative control, and impro...

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...

Simulation capacity is essential to the engineering design of industrial products with complex functions. This paper discusses a numerical algorithm to simulate the multiscale heat and moisture transfer in the thermal smart clothing system. A group of multiscale nonlinear models are proposed to describe the mix-type coupled heat and moisture transf...

In spectral decomposition of a 3D mesh model, it is well known that eigenvectors with respect to small eigenvalues determine its main pose while eigenvectors associated with large eigenvalues encode its surface details. Based on this property, given two meshes with different connectivities, one can use coupled quasi-harmonic technique to transfer t...

Currently most quad mesh simplification and optimization methods are based on the extension of triangle mesh simplification and optimization locally. The quality of the structure of a quad mesh is affected by its helical strips, and adjusting these helical strips can significantly improve the quality of the quad mesh in topological structure. In pa...

Compatible triangulation of two planar polygonal regions is an important operation for shape morphing and interpolation. This paper proposes an efficient algorithm to compatibly triangulate two given isomorphic planar polygonal shapes, which alternatively performs two operations, compatible triangulation and deformed energy minimization. The first...

Furniture arrangement is an important part of virtual scene design, which can be widely applied to virtual reality, 3D games and interior design. Noticing that existing methods for placing furniture are prone to converge to local optimal solutions because of constraint confliction among cost terms, we present a hierarchical optimization strategy an...

Cage-based deformation has become a popular method for shape deformation in computer graphics and animation. To edit a shape first requires a cage to be built to envelop the target model which is a tedious work by manual approaches. In this paper, we develop an automatic method to generate the cage for a model using voxelization based decomposition...

Acquisition of dynamic geometry is easier today. This makes compact and editable representations of source animated meshes more and more fascinating. Though several cage-based methods have been developed for reverse engineering of animated 3D models, none of them is truly error controllable. We address the problem by introducing a framework which a...

Shape interpolation, as a bridge communicating static geometries and dynamic shape sequences, is a fundamental operation in digital geometry processing and computer animation. We propose a fast as-isometric-as-possible (AIAP) 3D mesh interpolation approach which casts the shape interpolation problem to finding an AIAP motion trajectory from the sta...

This paper proposes a framework for matting identification photos automatically. We firstly adopt image processing operations such as skin detection, k-means clustering, Grab cut segmentation and canny operator to generate a trimap with three regions, i.e., background, foreground and unknown regions. Then, we improve the Bayesian matting by introdu...

This paper proposes a conceptually simple strategy for adaptive subdivision of unified subdivision surfaces with sharp features. A key feature of the proposed adaptive subdivision algorithms is that it follows exactly the same rule of the respective subdivision, which ensures correct evaluation of both refined control vertices and the corresponding...

A novel approach is presented for interpolating two planar shapes using a shape sequence such that one shape is deformed into the other in an as-isometric-as-possible (AIAP) manner. The two shapes can have arbitrarily connected edge soup as long as they have the same connectivity. The interpolation is described by a nonlinear optimization problem t...

We introduce L1-medial skeleton as a curve skeleton representation for 3D point cloud data. The L1-median is well-known as a robust global center of an arbitrary set of points. We make the key observation that adapting L1-medians locally to a point set representing a 3D shape gives rise to a one-dimensional structure, which can be seen as a localiz...