Charlie C. L. Wang

Neelotpal Dutta, Tianyu Zhang, Tao Liu et al.

Existing curved-layer-based process planning methods for multi-axis manufacturing address collisions only indirectly and generate toolpaths in a post-processing step, leaving toolpath geometry uncontrolled during optimization. We present an implicit neural field-based framework for multi-axis process planning that overcomes these limitations by embedding both layer generation and toolpath design within a single differentiable pipeline. Using sinusoidally activated neural networks to represent layers and toolpaths as implicit fields, our method enables direct evaluation of field values and derivatives at any spatial point, thereby allowing explicit collision avoidance and joint optimization of manufacturing layers and toolpaths. We further investigate how network hyperparameters and objective definitions influence singularity behavior and topology transitions, offering built-in mechanisms for regularization and stability control. The proposed approach is demonstrated on examples in both additive and subtractive manufacturing, validating its generality and effectiveness.

Wenting Wang, Zhuo Huang, Kun Qian et al.

We present FrameTwin, a curve-anchored Gaussian alignment framework that uses sparse-view images to close the control loop for adaptive wireframe 3D printing. Our key idea is to capture the deformation of thin wireframe structures from sparse-view images using Gaussian kernels anchored to parametric curves, yielding a compact and geometry-aware encoding that explicitly captures strut topology. Driven by a differentiable rendering pipeline, FrameTwin estimates a neural deformation field that aligns the partially printed target model with the deformed structure observed during fabrication, where the optimized curve-Gaussian representation serves as a digital twin of the evolving wireframe. Unlike general Gaussian-splatting approaches, our formulation constrains kernel placement along parametric curves, substantially reducing the ambiguity inherent in sparse-view observations of thin structures. The resultant deformation-field alignment enforces global consistency across all struts. By using the estimated deformation field to blend the distorted printed geometry with the remaining unprinted geometry, FrameTwin enables adaptive updates to future printing trajectories. We demonstrate that FrameTwin can robustly capture and compensate for deformation in wireframe models fabricated using a robotized 3D printing system.

Mian Qin, Junhao Ding, Shuo Qu et al.

Laser powder bed fusion (LPBF) is a widely used metal additive manufacturing technology. However, the accumulation of internal residual stress during printing can cause significant distortion and potential failure. Although various scan patterns have been studied to reduce possible accumulated stress, such as zigzag scanning vectors with changing directions or a chessboard-based scan pattern with divided small islands, most conventional scan patterns cannot significantly reduce residual stress. The proposed adaptive toolpath generation (ATG) algorithms, aiming to minimize the thermal gradients, may result in extremely accumulated temperature fields in some cases. To address these issues, we developed a deep reinforcement learning (DRL)-based toolpath generation framework, with the goal of achieving uniformly distributed heat and avoiding extremely thermal accumulation regions during the LPBF process. We first developed an overall pipeline for the DRL-based toolpath generation framework, which includes uniformly sampling, agent moving and environment observation, action selection, moving constraints, rewards calculation, and the training process. To accelerate the training process, we simplified the data-intensive numerical model by considering the turning angles on the toolpath. We designed the action spaces with three options, including the minimum temperature value, the smoothest path, and the second smoothest path. The reward function was designed to minimize energy density to ensure the temperature field remains relatively stable. To verify the effectiveness of the proposed DRL-based toolpath generation framework, we performed numerical simulations of polygon shape printing domains. In addition, four groups of thin plate samples with different scan patterns were compared using the LPBF process.

Tao Liu, Tianyu Zhang, Yongxue Chen et al.

We propose a neural network-based computational framework for the simultaneous optimization of structural topology, curved layers, and path orientations to achieve strong anisotropic strength in fiber-reinforced thermoplastic composites while ensuring manufacturability. Our framework employs three implicit neural fields to represent geometric shape, layer sequence, and fiber orientation. This enables the direct formulation of both design and manufacturability objectives - such as anisotropic strength, structural volume, machine motion control, layer curvature, and layer thickness - into an integrated and differentiable optimization process. By incorporating these objectives as loss functions, the framework ensures that the resultant composites exhibit optimized mechanical strength while remaining its manufacturability for filament-based multi-axis 3D printing across diverse hardware platforms. Physical experiments demonstrate that the composites generated by our co-optimization method can achieve an improvement of up to 33.1% in failure loads compared to composites with sequentially optimized structures and manufacturing sequences.

Bo Xu, Yuhu Guo, Yuchao Wang et al.

We propose a semantic-aware neural reconstruction method to generate 3D high-fidelity models from sparse images. To tackle the challenge of severe radiance ambiguity caused by mismatched features in sparse input, we enrich neural implicit representations by adding patch-based semantic logits that are optimized together with the signed distance field and the radiance field. A novel regularization based on the geometric primitive masks is introduced to mitigate shape ambiguity. The performance of our approach has been verified in experimental evaluation. The average chamfer distances of our reconstruction on the DTU dataset can be reduced by 44% for SparseNeuS and 20% for VolRecon. When working as a plugin for those dense reconstruction baselines such as NeuS and Neuralangelo, the average error on the DTU dataset can be reduced by 69% and 68% respectively.

Yabin Xu, Liangliang Nan, Laishui Zhou et al.

Reconstruction of high-fidelity 3D objects or scenes is a fundamental research problem. Recent advances in RGB-D fusion have demonstrated the potential of producing 3D models from consumer-level RGB-D cameras. However, due to the discrete nature and limited resolution of their surface representations (e.g., point- or voxel-based), existing approaches suffer from the accumulation of errors in camera tracking and distortion in the reconstruction, which leads to an unsatisfactory 3D reconstruction. In this paper, we present a method using on-the-fly implicits of Hermite Radial Basis Functions (HRBFs) as a continuous surface representation for camera tracking in an existing RGB-D fusion framework. Furthermore, curvature estimation and confidence evaluation are coherently derived from the inherent surface properties of the on-the-fly HRBF implicits, which devote to a data fusion with better quality. We argue that our continuous but on-the-fly surface representation can effectively mitigate the impact of noise with its robustness and constrain the reconstruction with inherent surface smoothness when being compared with discrete representations. Experimental results on various real-world and synthetic datasets demonstrate that our HRBF-fusion outperforms the state-of-the-art approaches in terms of tracking robustness and reconstruction accuracy.

Tianyu Zhang, Xiangjia Chen, Guoxin Fang et al.

Multi-axis additive manufacturing enables high flexibility of material deposition along dynamically varied directions. The Cartesian motion platforms of these machines include three parallel axes and two rotational axes. Singularity on rotational axes is a critical issue to be tackled in motion planning for ensuring high quality of manufacturing results. The highly nonlinear mapping in the singular region can convert a smooth toolpath with uniformly sampled waypoints defined in the model coordinate system into a highly discontinuous motion in the machine coordinate system, which leads to over-extrusion / under-extrusion of materials in filament-based additive manufacturing. The problem is challenging as both the maximal and the minimal speeds at the tip of a printer head must be controlled in motion. Moreover, collision may occur when sampling-based collision avoidance is employed. In this paper, we present a motion planning method to support the manufacturing realization of designed toolpaths for multi-axis additive manufacturing. Problems of singularity and collision are considered in an integrated manner to improve the motion therefore the quality of fabrication.

Guoxin Fang, Yingjun Tian, Zhi-Xin Yang et al.

This paper presents an efficient learning-based method to solve the inverse kinematic (IK) problem on soft robots with highly non-linear deformation. The major challenge of efficiently computing IK for such robots is due to the lack of analytical formulation for either forward or inverse kinematics. To address this challenge, we employ neural networks to learn both the mapping function of forward kinematics and also the Jacobian of this function. As a result, Jacobian-based iteration can be applied to solve the IK problem. A sim-to-real training transfer strategy is conducted to make this approach more practical. We first generate a large number of samples in a simulation environment for learning both the kinematic and the Jacobian networks of a soft robot design. Thereafter, a sim-to-real layer of differentiable neurons is employed to map the results of simulation to the physical hardware, where this sim-to-real layer can be learned from a very limited number of training samples generated on the hardware. The effectiveness of our approach has been verified on pneumatic-driven soft robots for path following and interactive positioning.

Rob B. N. Scharff, Guoxin Fang, Yingjun Tian et al.

Real-time proprioception is a challenging problem for soft robots, which have almost infinite degrees-of-freedom in body deformation. When multiple actuators are used, it becomes more difficult as deformation can also occur on actuators caused by interaction between each other. To tackle this problem, we present a method in this paper to sense and reconstruct 3D deformation on pneumatic soft robots by first integrating multiple low-cost sensors inside the chambers of pneumatic actuators and then using machine learning to convert the captured signals into shape parameters of soft robots. An exterior motion capture system is employed to generate the datasets for both training and testing. With the help of good shape parameterization, the 3D shape of a soft robot can be accurately reconstructed from signals obtained from multiple sensors. We demonstrate the effectiveness of this approach on two designs of soft robots -- a robotic joint and a deformable membrane. After parameterizing the deformation of these soft robots into compact shape parameters, we can effectively train the neural networks to reconstruct the 3D deformation from the sensor signals. The sensing and shape prediction pipeline can run at 50Hz in real-time on a consumer-level device.

Chuhua Xian, Kun Qian, Zitian Zhang et al.

Limited by the cost and technology, the resolution of depth map collected by depth camera is often lower than that of its associated RGB camera. Although there have been many researches on RGB image super-resolution (SR), a major problem with depth map super-resolution is that there will be obvious jagged edges and excessive loss of details. To tackle these difficulties, in this work, we propose a multi-scale progressive fusion network for depth map SR, which possess an asymptotic structure to integrate hierarchical features in different domains. Given a low-resolution (LR) depth map and its associated high-resolution (HR) color image, We utilize two different branches to achieve multi-scale feature learning. Next, we propose a step-wise fusion strategy to restore the HR depth map. Finally, a multi-dimensional loss is introduced to constrain clear boundaries and details. Extensive experiments show that our proposed method produces improved results against state-of-the-art methods both qualitatively and quantitatively.

Tim Kuipers, Eugeni L. Doubrovski, Jun Wu et al.

3D printing techniques such as Fused Deposition Modeling (FDM) have enabled the fabrication of complex geometry quickly and cheaply. High stiffness parts are produced by filling the 2D polygons of consecutive layers with contour-parallel extrusion toolpaths. Uniform width toolpaths consisting of inward offsets from the outline polygons produce over- and underfill regions in the center of the shape, which are especially detrimental to the mechanical performance of thin parts. In order to fill shapes with arbitrary diameter densely the toolpaths require adaptive width. Existing approaches for generating toolpaths with adaptive width result in a large variation in widths, which for some hardware systems is difficult to realize accurately. In this paper we present a framework which supports multiple schemes to generate toolpaths with adaptive width, by employing a function to decide the number of beads and their widths. Furthermore, we propose a novel scheme which reduces extreme bead widths, while limiting the number of altered toolpaths. We statistically validate the effectiveness of our framework and this novel scheme on a data set of representative 3D models, and physically validate it by developing a technique, called back pressure compensation, for off-the-shelf FDM systems to effectively realize adaptive width.

Chenming Wu, Yong-Jin Liu, Charlie C. L. Wang

Multi-directional 3D printing has the capability of decreasing or eliminating the need for support structures. Recent work proposed a beam-guided search algorithm to find an optimized sequence of plane-clipping, which gives volume decomposition of a given 3D model. Different printing directions are employed in different regions to fabricate a model with tremendously less support (or even no support in many cases).To obtain optimized decomposition, a large beam width needs to be used in the search algorithm, leading to a very time-consuming computation. In this paper, we propose a learning framework that can accelerate the beam-guided search by using a smaller number of the original beam width to obtain results with similar quality. Specifically, we use the results of beam-guided search with large beam width to train a scoring function for candidate clipping planes based on six newly proposed feature metrics. With the help of these feature metrics, both the current and the sequence-dependent information are captured by the neural network to score candidates of clipping. As a result, we can achieve around 3x computational speed. We test and demonstrate our accelerated decomposition on a large dataset of models for 3D printing.

Chuhua Xian, Dongjiu Zhang, Chengkai Dai et al.

Using the raw data from consumer-level RGB-D cameras as input, we propose a deep-learning based approach to efficiently generate RGB-D images with completed information in high resolution. To process the input images in low resolution with missing regions, new operators for adaptive convolution are introduced in our deep-learning network that consists of three cascaded modules -- the completion module, the refinement module and the super-resolution module. The completion module is based on an architecture of encoder-decoder, where the features of input raw RGB-D will be automatically extracted by the encoding layers of a deep neural-network. The decoding layers are applied to reconstruct the completed depth map, which is followed by a refinement module to sharpen the boundary of different regions. For the super-resolution module, we generate RGB-D images in high resolution by multiple layers for feature extraction and a layer for up-sampling. Benefited from the adaptive convolution operators newly proposed in this paper, our results outperform the existing deep-learning based approaches for RGB-D image complete and super-resolution. As an end-to-end approach, high fidelity RGB-D images can be generated efficiently at the rate of around 21 frames per second.

Bin Liu, Xiuping Liu, Zhixin Yang et al.

In this paper, we revisit the problem of 3D human modeling from two orthogonal silhouettes of individuals (i.e., front and side views). Different from our prior work, a supervised learning approach based on convolutional neural network (CNN) is investigated to solve the problem by establishing a mapping function that can effectively extract features from two silhouettes and fuse them into coefficients in the shape space of human bodies. A new CNN structure is proposed in our work to exact not only the discriminative features of front and side views and also their mixed features for the mapping function. 3D human models with high accuracy are synthesized from coefficients generated by the mapping function. Existing CNN approaches for 3D human modeling usually learn a large number of parameters (from 8.5M to 355.4M) from two binary images. Differently, we investigate a new network architecture and conduct the samples on silhouettes as input. As a consequence, more accurate models can be generated by our network with only 2.4M coefficients. The training of our network is conducted on samples obtained by augmenting a publicly accessible dataset. Learning transfer by using datasets with a smaller number of scanned models is applied to our network to enable the function of generating results with gender-oriented (or geographical) patterns.

Chengkai Dai, Sylvain Lefebvre, Kai-Ming Yu et al.

We present a method for effectively planning the motion trajectory of robots in manufacturing tasks, the tool-paths of which are usually complex and have a large number of discrete-time constraints as waypoints. Kinematic redundancy also exists in these robotic systems. The jerk of motion is optimized in our trajectory planning method at the meanwhile of fabrication process to improve the quality of fabrication.

Chenming Wu, Chengkai Dai, Guoxin Fang et al.

We present a method for fabricating general models with multi-directional 3D printing systems by printing different model regions along with different directions. The core of our method is a support-effective volume decomposition algorithm that minimizes the area of the regions with large overhangs. A beam-guided searching algorithm with manufacturing constraints determines the optimal volume decomposition, which is represented by a sequence of clipping planes. While current approaches require manually assembling separate components into a final model, our algorithm allows for directly printing the final model in a single pass. It can also be applied to models with loops and handles. A supplementary algorithm generates special supporting structures for models where supporting structures for large overhangs cannot be eliminated. We verify the effectiveness of our method using two hardware systems: a Cartesian-motion based system and an angular-motion based system. A variety of 3D models have been successfully fabricated on these systems.

Shuo Jin, Chengkai Dai, Yang Liu et al.

Existing techniques for motion imitation often suffer a certain level of latency due to their computational overhead or a large set of correspondence samples to search. To achieve real-time imitation with small latency, we present a framework in this paper to reconstruct motion on humanoids based on sparsely sampled correspondence. The imitation problem is formulated as finding the projection of a point from the configuration space of a human's poses into the configuration space of a humanoid. An optimal projection is defined as the one that minimizes a back-projected deviation among a group of candidates, which can be determined in a very efficient way. Benefited from this formulation, effective projections can be obtained by using sparse correspondence. Methods for generating these sparse correspondence samples have also been introduced. Our method is evaluated by applying the human's motion captured by a RGB-D sensor to a humanoid in real-time. Continuous motion can be realized and used in the example application of tele-operation.

Gang Xu, Tsz-Ho Kwok, Charlie C. L. Wang

Volumetric spline parameterization and computational efficiency are two main challenges in isogeometric analysis (IGA). To tackle this problem, we propose a framework of computation reuse in IGA on a set of three-dimensional models with similar semantic features. Given a template domain, B-spline based consistent volumetric parameterization is first constructed for a set of models with similar semantic features. An efficient quadrature-free method is investigated in our framework to compute the entries of stiffness matrix by Bezier extraction and polynomial approximation. In our approach, evaluation on the stiffness matrix and imposition of the boundary conditions can be pre-computed and reused during IGA on a set of CAD models. Examples with complex geometry are presented to show the effectiveness of our methods, and efficiency similar to the computation in linear finite element analysis can be achieved for IGA taken on a set of models.