To meet the fashion industry’s increasing demand for intelligent tools capable of manipulating and retexturing fashion items while maintaining structural integrity, an area where existing methods often lack photorealistic quality and intuitive control, we introduce iRetexturing, a novel framework that integrates diffusion models and geometric priors to tackle these challenges. The iRetexturing method utilizes a three-stage pipeline: high-resolution preprocessing with masked super-resolution and semantic segmentation, quadripartite texture synthesis using grid-based tiling and boundary-aware regeneration, and ControlNet-guided diffusion with dual spatial constraints (Canny edges and depth maps), incorporating innovations such as parametric texture modulation, partial repainting with seam refinement, and real-time adaptability across diverse materials. Evaluations conducted on 4400 fashion images reveal that iRetexturing outperforms state-of-the-art methods including diffuseIT, achieving a learned perceptual image patch similarity of 0.1385 and structural similarity of 0.8323 compared with diffuseIT’s 0.1618 and 0.8074, respectively, despite diffuseIT’s lower Fréchet inception distance (54.96 versus 75.25), highlighting iRetexturing’s superiority in fine-grained texture replacement and high-fidelity textile design. By combining diffusion-based generation with geometric priors, iRetexturing enables precise manipulation of surface characteristics, bridging the gap between conceptual prototypes and production-ready assets, and offering transformative potential for the fashion industry by streamlining the design-to-production process and fostering creative innovation.
Many man-made or natural objects are composed of symmetric parts and possess symmetric physical behavior. Although its shape can exactly follow a symmetry in the designing or modeling stage, its discretized mesh in the analysis stage may be asymmetric because generating a mesh exactly following the symmetry is usually costly. As a consequence, the expected symmetric physical behavior may not be faithfully reproduced due to the asymmetry of the mesh. To solve this problem, we propose to optimize the material parameters of the mesh for static and kinematic symmetry behavior. Specifically, under the situation of static equilibrium, Young’s modulus is properly scaled so that a symmetric force field leads to symmetric displacement. For kinematics, the mass is optimized to reproduce symmetric acceleration under a symmetric force field. To efficiently measure the deviation from symmetry, we formulate a linear operator whose kernel contains all the symmetric vector fields, which helps to characterize the asymmetry error via a simple norm. To make the resulting material suitable for the general situation, the symmetric training force fields are derived from modal analysis in the above kernel space. Results show that our optimized material significantly reduces the asymmetric error on an asymmetric mesh in both static and dynamic simulations.
With the explosive growth of 3D objects yet expensive annotation costs, unsupervised 3D object retrieval has become a popular but challenging research area. Existing labeled resources have been utilized to aid this task via transfer learning, which aligns the distribution of unlabeled data with the source one. However, the labeled resource are not always accessible due to the privacy disputes, limited computational capacity and other thorny restrictions. Therefore, we propose source-free model adaptation task for unsupervised 3D object management, which utilizes a pre-trained model to boost the performance with no access to source data and labels. Specifically, we compute representative prototypes to assume the source feature distribution, and design a bidirectional cumulative confidence-based adaptation strategy to adaptively align unlabeled samples towards prototypes. Subsequently, a dual-model distillation mechanism is proposed to generate source hypothesis for remedying the absence of ground-truth labels. The experiments on a cross-domain retrieval benchmark NTU-PSB (PSB-NTU) and a cross-modality retrieval benchmark MI3DOR also demonstrate the superiority of the proposed method even without access to raw data.