While recent advances in Gaussian Splatting have enabled fast reconstruction of high-quality 3D scenes from images, extracting accurate surface meshes remains a challenge. Current approaches extract the surface through costly post-processing steps, resulting in the loss of fine geometric details or requiring significant time and leading to very dense meshes with millions of vertices. More fundamentally, the a posteriori conversion from a volumetric to a surface representation limits the ability of the final mesh to preserve all geometric structures captured during training. We present MILo, a novel Gaussian Splatting framework that bridges the gap between volumetric and surface representations by differentiably extracting a mesh from the 3D Gaussians. We design a fully differentiable procedure that constructs the mesh—including both vertex locations and connectivity—at every iteration directly from the parameters of the Gaussians, which are the only quantities optimized during training. Our method introduces three key technical contributions: (1) a bidirectional consistency framework ensuring both representations—Gaussians and the extracted mesh—capture the same underlying geometry during training; (2) an adaptive mesh extraction process performed at each training iteration, which uses Gaussians as differentiable pivots for Delaunay triangulation; (3) a novel method for computing signed distance values from the 3D Gaussians that enables precise surface extraction while avoiding geometric erosion. Our approach can reconstruct complete scenes, including backgrounds, with state-of-the-art quality while requiring an order of magnitude fewer mesh vertices than previous methods. Due to their light weight and empty interior, our meshes are well suited for downstream applications such as physics simulations and animation. The code for our approach and an online gallery are available at https://anttwo.github.io/milo/.

Diffusion models have significantly advanced image manipulation techniques, and their ability to generate photorealistic images is beginning to transform retail workflows, particularly in presale visualization. Beyond artistic style transfer, the capability to perform fine-grained visual feature transfer is becoming increasingly important. Embroidery is a textile art form characterized by intricate interplay of diverse stitch patterns and material properties, which poses unique challenges for existing style transfer methods. To explore the customization for such fine-grained features, we propose a novel contrastive learning framework that disentangles fine-grained style and content features with a single reference image, building on the classic concept of image analogy. We first construct an image pair to define the target style, and then adopt a similarity metric based on the decoupled representations of pretrained diffusion models for style-content separation. Subsequently, we propose a two-stage contrastive LoRA modulation technique to capture fine-grained style features. In the first stage, we iteratively update the whole LoRA and the selected style blocks to initially separate style from content. In the second stage, we design a contrastive learning strategy to further decouple style and content through self-knowledge distillation. Finally, we build an inference pipeline to handle image or text inputs with only the style blocks. To evaluate our method on fine-grained style transfer, we build a benchmark for embroidery customization. Our approach surpasses prior methods on this task and further demonstrates strong generalization to three additional domains: artistic style transfer, sketch colorization, and appearance transfer. Our project is available at: https://style3d.github.io/embroidery_customization.

Neural implicit shape representation has drawn significant attention in recent years due to its smoothness, differentiability, and topological flexibility. However, directly modeling the shape of a neural implicit surface, especially as the zero-level set of a neural signed distance function (SDF), with sparse geometric control is still a challenging task. Sparse input shape control typically includes 3D curve networks or, more generally, 3D curve sketches, which are unstructured and cannot be connected to form a curve network, and therefore more difficult to deal with. While 3D curve networks or curve sketches provide intuitive shape control, their sparsity and varied topology pose challenges in generating high-quality surfaces to meet such curve constraints. In this paper, we propose NeuVAS, a variational approach to shape modeling using neural implicit surfaces constrained under sparse input shape control, including unstructured 3D curve sketches as well as connected 3D curve networks. Specifically, we introduce a smoothness term based on a functional of surface curvatures to minimize shape variation of the zero-level set surface of a neural SDF. We also develop a new technique to faithfully model G 0 sharp feature curves as specified in the input curve sketches. Comprehensive comparisons with the state-of-the-art methods demonstrate the significant advantages of our method.

Remanufacturing effectively extends component lifespans by restoring used or end-of-life parts to like-new or even superior conditions, with an emphasis on maximizing reutilized material, especially for high-cost materials. Hybrid manufacturing technology combines the capabilities of additive and subtractive manufacturing, with the ability to add and remove material, enabling it to remanufacture complex shapes and is increasingly being applied in remanufacturing. How to effectively plan the process of additive and subtractive hybrid remanufacturing (ASHRM) to maximize material reutilization has become a key focus of attention. However, current ASHRM process planning methods lack strict consideration of collision-free constraints, hindering practical application. This paper introduces a computational framework to tackle ASHRM process planning for general shapes with strictly considering these constraints. We separate global and local collision-free constraints, employing clipping planes and graph to tackle them respectively, ultimately maximizing the reutilized volume while ensuring these constraints are satisfied. Additionally, we also optimize the setup of the target model that is conducive to maximizing the reutilized volume. Extensive experiments and physical validations on a 5-axis hybrid manufacturing platform demonstrate the effectiveness of our method across various 3D shapes, achieving an average material reutilization of 69% across 12 cases. Code is publicly available at https://github.com/fanchao98/Waste-to-Value.

The detection and computation of the overlap region between two NURBS surfaces, as a special case of the intersection problem, are essential components of CAD systems, directly influencing the robustness of the entire system. Despite their importance, efficient, topologically correct, and numerically robust algorithms for detecting overlap regions remain lacking. To address this issue, we propose an optimization approach for computing the overlap region between two NURBS surfaces within a given error threshold. Based on a bilevel optimization framework, our algorithm first employs cubic Bézier simplices to approximate the boundary of the overlap region. The boundary points of the overlap region are computed iteratively, followed by a Delaunay triangulation to establish the boundary topology. Additional refinement of the boundary edge is applied to ensure the topological correctness and maintain the precision of the overlap region within the specified error threshold. Our main contribution lies in the development of a novel and robust algorithm to calculate the boundary of the overlap region. This approach differs from previous overlap computation methods, which seldom account for error thresholds and are difficult to implement in floating-point arithmetic in CAD systems. We demonstrate the robustness and topological accuracy of our method through extensive experiments on a diverse set of complex examples with varying error thresholds.

Selection is the first step in many image editing processes, enabling faster and simpler modifications of all pixels sharing a common modality. In this work, we present a method for material selection in images, robust to lighting and reflectance variations, which can be used for downstream editing tasks. We rely on vision transformer (ViT) models and leverage their features for selection, proposing a multi-resolution processing strategy that yields finer and more stable selection results than prior methods. Furthermore, we enable selection at two levels: texture and subtexture, leveraging a new two-level material selection (DuMaS) dataset which includes dense annotations for over 800,000 synthetic images, both on the texture and subtexture levels.

We present a high-speed underwater optical backscatter communication technique based on acousto-optic light steering. Our approach enables underwater assets to transmit data at rates potentially reaching hundreds of Mbps, vastly outperforming current state-of-the-art optical and underwater backscatter systems, which typically operate at only a few kbps. In our system, a base station illuminates the backscatter device with a pulsed laser and captures the retroreflected signal using an ultrafast photodetector. The backscatter device comprises a retroreflector and a 2 MHz ultrasound transducer. The transducer generates pressure waves that dynamically modulate the refractive index of the surrounding medium, steering the light either toward the photodetector (encoding bit 1) or away from it (encoding bit 0). Using a 3-bit redundancy scheme, our prototype achieves a communication rate of approximately 0.66 Mbps with an energy consumption of ≤ 1 μJ/bit, representing a 60× improvement over prior techniques. We validate its performance through extensive laboratory experiments in which remote underwater assets wirelessly transmit multimedia data to the base station under various environmental conditions.

This paper presents LEGO ® -Maker, a new learning-based generative model that can effectively consider over 100 unique brick types and rapidly generate hundreds of bricks to create LEGO ® models conditioned on images. This work has three major technical contributions that enable it to achieve surpassing capabilities beyond existing generative approaches. First, we design a compact LEGO ® tokenization scheme to serialize LEGO ® models and bricks into tokens for autoregressive learning. Second, we build LEGO ® -Maker, an autoregressive image-conditioned architecture, with a multi-token prediction strategy to encourage pre-considering multiple brick attributes and a rollback mechanism for collision-free generation. Third, we propose an effective data preparation pipeline with a procedural generator to synthesize LEGO ® models and a LEGO ® -to-real image translator distilled from a large vision language model to translate LEGO ® renderings into associated photorealistic images, leveraging rich prior to address the scarcity of image-to-LEGO ® data. Extensive evaluations and comparisons are conducted on two object categories, facade and portrait, over metrics in four aspects: geometry, color, semantics, and structural integrity, together with a user study. Experimental results demonstrate the versatility and compelling strengths of LEGO ® -Maker in producing structures and details given by the reference image. Also, the evaluation scores manifest that our method clearly surpasses the baselines, consistently for all evaluation metrics.

We introduce AnySplat, a feed-forward network for novel-view synthesis from uncalibrated image collections. In contrast to traditional neural-rendering pipelines that demand known camera poses and per-scene optimization, or recent feed-forward methods that buckle under the computational weight of dense views—our model predicts everything in one shot. A single forward pass yields a set of 3D Gaussian primitives encoding both scene geometry and appearance, and the corresponding camera intrinsics and extrinsics for each input image. This unified design scales effortlessly to casually captured, multi-view datasets without any pose annotations. In extensive zero-shot evaluations, AnySplat matches the quality of pose-aware baselines in both sparse- and dense-view scenarios while surpassing existing pose-free approaches. Moreover, it greatly reduces rendering latency compared to optimization-based neural fields, bringing real-time novel-view synthesis within reach for unconstrained capture settings. Project page: https://city-super.github.io/anysplat/.

Animation retargetting applies sparse motion description (e.g., keypoint sequences) to a character mesh to produce a semantically plausible and temporally coherent full-body mesh sequence. Existing approaches come with restrictions - they require access to template-based shape priors or artist-designed deformation rigs, suffer from limited generalization to unseen motion and/or shapes, or exhibit motion jitter. We propose Self-supervised Motion Fields (SMF), a self-supervised framework that is trained with only sparse motion representations, without requiring dataset-specific annotations, templates, or rigs. At the heart of our method are Kinetic Codes, a novel autoencoder-based sparse motion encoding, that exposes a semantically rich latent space, simplifying large-scale training. Our architecture comprises dedicated spatial and temporal gradient predictors, which are jointly trained in an end-to-end fashion. The combined network, regularized by the Kinetic Codes' latent space, has good generalization across both unseen shapes and new motions. We evaluated our method on unseen motion sampled from AMASS, D4D, Mixamo, and raw monocular video for animation transfer on various characters with varying shapes and topology. We report a new SoTA on the AMASS dataset in the context of generalization to unseen motion.