Dejun Zheng

Fuzzy region competition-based auto-color-theme design for textile images

We propose a model to recolor textile images by different color themes. The model contains three phases. The first phase is to partition an input textile image into several homogeneous regions. The CIELab color mean of each region and a bias-field function are obtained from the segmentation results. The combination of the color mean values of all regions is considered as the color theme of the input image. The second phase is to retrieve the relevant color themes from a given dataset. The retrieved color themes preserve the color mood of the input image in the sense of the similarity measurement defined in the color mood space. In the third phase, we reconstruct new images with different appearances from the input image by using the retrieved color themes. The proposed method provides a powerful tool for designers to generate and search for all relevant color combinations related to a given theme. Numerical results indicate that our recolorization model performs well on complex textile design patterns.

Partial matching of garment panel shapes with dynamic sketching design

Fashion industry and textile manufacturing in past decade, have been starting to reapply enhanced intelligent CAD process technologies. In this paper, we propose a partial panel matching system to facilitate the typical garment design process. This process provides recommendations to the designer during the panel design process and performs partial matching of the garment panel shapes. There are three main parts in our partial matching system. First, we make use of a Bézier-based sketch regularization to pre-process the panel sketch data. Second, we propose a set of bi-segment panel shape descriptors to describe and enrich the local features of the shape for partial matching. Finally, based on our previous work, we add an interactive sketching input environment to design garments. Experiment results show the effectiveness and efficiency of the proposed system.

Accurate indexing and classification for fabric weave patterns using entropy-based approach

In current textile design, fabric weave pattern indexing and searching require extensive manual operations. The manual weave pattern classification is not sufficient to give the accurate and precise result and it is time-consuming. There is no such research to index and search for weave pattern specially. In this paper we propose a method to index and search weave patterns. We use pattern clusters, transitions, entropy and Fast Fourier Transform (FFT) directionality as a hybrid approach for the cognitive comparison and classification of weave pattern. There are three common patterns used in textile design. They are plain weave, twill weave and satin weave patterns. First, we classify weave patterns into these three categories according to weave pattern definition and weave point distribution characteristics (weave pattern smoothness and connectivity). Second, we use the FFT to describe the weave point distribution. Finally, we use entropy method to calculate the weave point distribution into a significant index value. Our approach can avoid the problem of pattern duplications in the database. In our experiment, we select and test commonly used weave patterns with our proposed approach. Our experiment results show that our approach can achieve substantially accurate classification.

IDSS: A Novel Representation for Woven Fabrics

The appearance of woven fabrics is intrinsically determined by the geometric details of their meso/micro scale structure. In this paper, we propose a multiscale representation and tessellation approach for woven fabrics. We extend the Displaced Subdivision Surface (DSS) to a representation named Interlaced/Intertwisted Displacement Subdivision Surface (IDSS). IDSS maps the geometric detail, scale by scale, onto a ternary interpolatory subdivision surface that is approximated by Bezier patches. This approach is designed for woven fabric rendering on DX11 GPUs. We introduce the Woven Patch, a structure based on DirectXs new primitive, patch, to describe an area of a woven fabric so that it can be easily implemented in the graphics pipeline using a hull shader, a tessellator and a domain shader. We can render a woven piece of fabric at 25 frames per second on a low-performance NVIDIA 8400 MG mobile GPU. This allows for large-scale representations of woven fabrics that maintain the geometric variances of real yarn and fiber.

Design through cognitive color theme: A new approach for fabric design

Fashion designers conceive color composition as an ensemble of tones, shades and tints that often resemble an abstract or natural theme. Color and tone adjustments are among the most frequent fabric design operations. Color and tone style adjustments are part of a generic process of cognition involved in the creation of new fabric designs. A challenging problem in this process is given a s et of natural images associated with a subject or a theme, the problem is to discover the underlying cognitive relationships that connect textures and color tones that are perceived by an observer or in our case, a fabric designer. These relationships culminate in optimal combinations of colors and tones that resemble the original theme in which a template of colors is defined with respect to an associated verbal description. We formulate the fabric design process as a cognitive process that simultaneously considers multiple color tones to form a desired theme as well as texture characteristics in the fabric. In this paper, we incorporate color assembly and texture knowledge in the fabric design process in order to automatically produce the color combinations learnt from a given natural or abstract theme. We demonstrate that the use of color theme associations can automatically generate new fabric designs that rival complex commercial designs that are otherwise difficult to generate even by experienced designers. Preliminary experiments confirm the effectiveness of our method.

Cognitive weave pattern prioritization: An application-oriented approach

Very often, the recognition of a pattern is accompanied by a cognitive process of interpretation and understanding. In the arts and sciences, as well as in our daily lives, we learned patterns from nature and create new patterns for various applications. Weave pattern is one of the most important artificial patterns in our daily lives and there are numerous applications. To manipulate the weave patterns, texton indexing and prioritization are needed to perform, which is associated with a cognitive process of interpretation and understanding of pattern. In this regard, we use an interdisciplinary approach to help selecting weave texture patterns using tailored features and algorithms, taking into account essential features or rules of pattern design. The features and algorithms are designed based on the object-attribute-relation (OAR) model and cognitive informatics model. Three essential features of weave pattern are proposed, i.e. the complexity of patterns in production process, visual structural appearance and cognitive features to track for weave pattern. Our experiments on a wide variety of weave patterns show that the proposed approach is capable of effectively prioritizing weave texture patterns.