System and method for generating a zero-waste design pattern and reducing material waste

Abstract

The present invention discloses a system and method for generating a zero-waste design pattern, including but not limited to clothing, furniture, shoes, and other accessories, and reducing fabric or material waste. The method includes: (i) receiving a target design input including a first plurality of cut pieces; (ii) rendering a first 3D clothing surface from the first plurality of cut pieces of the target design input; (iii) repeatedly applying merging or splitting, optimization, and packing to the first plurality of cut pieces to generate a second plurality of cut pieces; (iv) rendering a second 3D clothing surface from the second plurality of cut pieces; and (v) comparing the first 3D clothing surface with the second 3D clothing surface, and repeatedly performing the task of applying merging or splitting, optimization, and packing when the deviation between the first 3D clothing surface and the second 3D clothing surface exceeds a predefined threshold.

Description

Technical Field

[0001] Preface In the following specification, the present invention and methods for implementing the present invention will be described in detail.

[0002] Description of the Present Invention The present invention generally relates to systems and methods for generating zero-waste design patterns, where the generated design patterns include, but are not limited to, clothing, furniture, shoes, and other accessories. More particularly, the present invention relates to a multi-task optimization method for achieving zero-waste design patterns and reducing fabric or any material waste during material processing.

Background Art

[0003] Generally, the most significant factor affecting the cost of clothing production is the fabric. However, in the fashion industry, 92 million tons of fabric are wasted every year. This problem is worsening. By 2030, it is expected that more than 134 million tons of fabric will be wasted every year. Fabric or material waste starts with design. For example, today, the clothing production process usually begins with the vision of a creative designer in the form of an abstract 2D illustration. Such illustrations aim at the aesthetics of the clothing but do not clarify the source of the fabric or how the fabric or material will ultimately be cut to become the final 3D item.

[0004] The creative illustration is then turned into a pattern, specification, and tech pack by a technical designer, pattern maker, or manufacturer who determines how the fabric or material is to be cut to achieve the original illustrated vision. The pattern, tech pack, and specification sheet consist of detailed information about the clothing design, including cut pieces, dimensions, care label instructions, artwork placement, fabric or material specifications, packing instructions, and other technical information about the product necessary to value and assemble the finished product. From the industry standard orientation that design is of utmost importance, the technical designer, pattern maker, or manufacturer works to get as close as possible to the original creative illustration, resulting in unevenly shaped cuts that prioritize the aesthetics of the set over any consideration of materials, generating fabric or material waste.

[0005] What complements manual design and pattern placement is to apply an automatic nesting algorithm that lays out the cut pieces in a way that reduces raw material waste. In clothing, the automatic nesting algorithm can reduce fabric or material waste by providing the placement of fabric or material pieces. However, since the fabric or material pieces remain in uneven shapes that cannot be changed, the pieces do not fit together, and fabric or material savings are limited to only a few percentage points. An exemplary automatic nesting algorithm achieves only a 4% fabric savings. Since most clothing and fabric-based furniture designs have 10 - 30% fabric and / or material waste, a 4% savings is far from zero waste.

[0006] Zero-waste design, or zero-fabric / material waste design, is a clothing design that does not produce any fabric or material scraps as waste. Zero waste can be achieved by strategically cutting, folding, or draping fabric or material. In the apparel space, although there are various forms of zero-waste clothing designs today, they mainly have the following drawbacks. (i) Zero-waste clothing designs consume more fabric or material through unnecessary folds, darts, or drapes, ultimately negating the material-saving effect associated with zero-waste design. (ii) Zero-waste clothing designs rely on complex or more cuts or patchwork. This increases the cutting time and / or sewing time, making the clothing more labor-intensive (e.g., due to an increase in cutting time) during the production process and thus more costly. (iii) Zero-waste clothing designs are limited to simple or "boxy" designs such as shifts and kimonos.

[0007] For example, U.S. Patent No. 10,588,369 (Patent Document 1) titled "Textile repurposing and sustainable garment design" discloses a method of upcycling multiple fabric articles to form clothing. In some embodiments, fabric or material pieces are positioned inside one or more pattern pieces such that their edges do not overlap and completely cover each piece. The positioned pieces are treated with an adhesive, and then a layer of paper is adhered thereto to hold them in place while the positioned fabric or material pieces are being sewn together. The resulting sandwich is then immersed in a liquid to remove the paper and dissolve the adhesive. The resulting integrated fabric or material component is then available for sewing with other similarly formed fabric components to form clothing. However, this prior art does not discuss (i) minimizing fabric or material consumption to enable total material savings, and (ii) minimizing the complexity of the clothing to reduce cutting and sewing times during production, thereby leading to a more efficient design.

[0008] Accordingly, there is a need for an interactive workflow that generates designs for clothing, footwear, accessories, and other items that achieve zero waste.

Prior Art Documents

Patent Documents

[0009]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0010] The present invention overcomes the drawbacks of the prior art by providing an interactive workflow that generates designs for clothing, shoes, accessories, furniture, and other items to achieve zero waste. The proposed method achieves zero waste while finding an optimized balance between (i) minimizing fabric or material consumption to enable total material savings (calculated as the difference in the length of fabric or area of material required compared to previous versions of the same or similar style), (ii) minimizing the complexity of the clothing to reduce cutting and sewing times during production and achieve a more efficient design, and (iii) closely approximating the original desired design pattern.

Means for Solving the Problems

[0011] According to an embodiment of the present disclosure, a method and a computer program product are provided for reducing fabric or material waste in material and accessory processing. A target design including a first plurality of cut pieces is considered. The first plurality of cut pieces is rendered as a first 3D surface. Merging or splitting, optimization, and packing are repeatedly applied to the first plurality of cut pieces to produce a second plurality of cut pieces. The second plurality of cut pieces is rendered as a second 3D surface. The first 3D surface and the second 3D surface are compared, and when the deviation between the first 3D surface and the second 3D surface exceeds a predetermined level, the merging or splitting, optimization, and packing are repeated.

[0012] In various embodiments, the present disclosure takes a fabric - first approach using two inputs: (i) the dimensions of the fabric or material, and (ii) the desired design style. This method then decomposes and combines the desired design within the creative constraints of the dimensions of the fabric or material. As a result, any creative design is forced to consider the use, shape, cut, and layout of the fabric or material so that the fabric or material pieces fit together without waste between them. The result is not the conventional asymmetrical / amorphous pieces, but fabric or material pieces like a puzzle that fit together. Although some of the outputted zero - waste cuts may be more linear, the output not only utilizes both sides of any curvilinear cuts for clothing design, but also makes good use of the flow of the soft fabric, thereby avoiding angular / constrained designs and looking forward to a silhouette of clothing that appears to fit the body without a sense of incongruity.

[0013] Embodiments of the present disclosure optimize shape, efficiency, and design while enabling zero waste. This leads to a significant improvement compared to other methods that conventionally discharge 10 - 30% of fabric or material waste for many clothing and interior decoration items. Clothing that requires additional fabric orientation or fabric print alignment (e.g., a comic print in the center of the front of a dress) may have a 40% increase in fabric or material waste. Skilled technical designers or pattern makers who must strictly adhere to the initial design illustration still leave about 15 - 20% of the fabric or material as waste on the cutting room floor. The process proposed by the present invention is not only limited to obtaining a zero - waste design pattern output, but also realizes a significant reduction in fabric or material consumption, often 25% to 30% or more, due to improved design efficiency.

[0014] The present invention provides a system for generating a zero-waste design pattern and reducing material waste. The system includes a computing node, and the computing node further includes a computer server capable of executing a process for reducing fabric or material waste in material processing. The computer server includes: (i) a system memory which is a computer-readable storage medium composed of a plurality of program modules for implementing a multi-task optimization method for reducing fabric or material waste in material processing; (ii) a plurality of processing units capable of executing the program modules stored in the system memory, the processing units sequentially executing a multi-task optimization method including steps of patch merging, patch shape optimization, strip packing, and patch splitting, which are repeatedly implemented to improve the packing efficiency of clothes; and (iii) a network adapter for enabling wired or wireless communication between components of the computer server via a bus.

[0015] Furthermore, the present invention provides a method for reducing fabric or material waste in clothing, accessories, and furniture manufacturing, the method comprising: (i) receiving a target design input comprising a first plurality of cut pieces including, but not limited to, a pattern, wherein according to one embodiment of the present invention, the target design input is not limited to cut pieces and also extends to input from a template library or a two-dimensional (2D) or three-dimensional (3D) design, the receiving step; (ii) rendering a first 3D clothing surface from the first plurality of cut pieces from the target design input; (iii) repeatedly applying merge or split, optimization, and packing to the first plurality of cut pieces including, but not limited to, a pattern to generate a second plurality of cut pieces; (iv) rendering a second 3D clothing surface from the second plurality of cut pieces including, but not limited to, a pattern; and (v) comparing the first 3D clothing surface and the second 3D clothing surface and repeatedly performing the task of applying merge, optimization, and packing when the deviation between the first 3D clothing surface and the second 3D clothing surface exceeds a predefined threshold.

[0016] In various embodiments, a conventional design tech pack, pattern, or design illustration in 2D or 3D is provided as input. Multifunctional optimization to the tech pack, pattern, or illustration is performed to find a locally optimal solution that (i) minimizes material waste and total fabric or material consumption based on fabric or material dimensions, (ii) minimizes cutting / sewing time, and (iii) maximizes similarity to the original design in 3D. In addition to optimization, the user can interact with the resulting 2D pattern or illustration to apply further aesthetic adjustments. User interaction and optimization can continue iteratively until a final pattern or tech pack is generated.

[0017] Embodiments of the present disclosure use an iterative process of interaction to (i) eliminate waste, (ii) simplify the making of clothing, and (iii) use the selected fabric or material as a creative constraint to retain the desired design for a desired clothing design, and then optimize the shape of the fabric or material piece (through patch shape optimization), optimize the placement (through a packing algorithm), and optimize the cut (through a combination of patch splitting or merging and shape optimization).

[0018] The above-described features and other features of the embodiments will become more apparent from the following detailed description of the embodiments when read in conjunction with the accompanying drawings. In the drawings, like reference numerals refer to like elements.

Brief Description of the Drawings

[0019]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 4a

Figure 4b

Figure 4c

Figure 4d

Figure 4e

Figure 4f

Figure 4g

Figure 4h

Figure 4i-1

Figure 4i-2

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10a

Figure 10b

Figure 11

Figure 12

[0020] Reference will now be made in detail to embodiments of the subject matter, one or more examples of which are illustrated in the figures. Each example is provided by way of explanation and not limitation. Various changes and modifications which will be apparent to those skilled in the art are considered to be within the spirit, scope, and intent of the invention.

[0021] FIG. 1 shows an exemplary jacket fabric cut. Examples of conventional jacket cuts (such as those included in patterns) are provided. The cuts include various irregular shapes, which significantly extend the cutting time and prevent an efficient layout on the fabric, contributing to waste.

[0022] FIG. 2 shows an exemplary jacket cut generated according to an embodiment of the present disclosure. The cut fits within a rectangular piece of fabric without any gaps, thereby minimizing both cutting time and waste. Additionally, by fitting all pattern pieces within the rectangular piece of fabric, each additional unit produced can be accurately placed along the edge of the previous unit, anticipating scalable production with zero fabric waste.

[0023] FIG. 3 shows the exemplary jacket of FIG. 2 in an assembled form. While having zero fabric waste, the edges are still positioned to fit the body.

[0024] Figure 4 shows a multi - task optimization method for reducing fabric or material waste in clothing production according to an embodiment of the present disclosure. In particular, the design style efficiently fits within the dimensions of the fabric or material with zero waste. Multi - objective optimization is performed using patch splitting or merging, patch shape optimization, and packing algorithms. For multiple objectives and given changing business objectives, the result may vary by (i) prioritizing the least deviation from the desired design while eliminating as much fabric or material waste as possible within the fabric or material dimensions, or (ii) prioritizing zero waste while allowing the design to be further adapted to fit within the fabric or material dimensions.

[0025] Referring to FIG. 4, a multi - task optimization method (400) for reducing fabric or material waste includes providing one or more inputs required for material processing to a system (100), where the provided inputs include material information (401), cut pieces (402) directly taken from a pattern / tech pack (403) or obtained from a 2D / 3D design (404) or design concept, and / or metadata indicating additional characteristics required for clothing production, where the metadata includes the clothing type (e.g., jacket, skirt) or the presence of additional features (e.g., pockets). In one embodiment, the material can include fabric, as well as leather, plastic, or other materials suitable for the production of clothing, shoes, accessories, and furniture, and the material information (401) can include material type, weight, dimensions, description, orientation based on texture direction or print, and print length. In the case of an unusual fabric (e.g., leather from invasive alien species made to support the ecosystem), in addition to the dimensions, a map of the fabric or material area may be included.

[0026] Templates, or multiple appropriate parts from multiple templates, are retrieved from a template library (405) based on the input provided, where the templates contain a set of default fabric pieces or default material pieces correlated with drawing metadata including, but not limited to, 2D / 3D sketches, patterns, or tech packs. For example, a template for a T-shirt with a chest pocket would include front and back pieces, two sleeve pieces, and a pocket piece. In some embodiments, the drawing and clothing type are provided to a learning system, which is pre-trained to output the dimensions to be applied to each of the cut pieces of the template to achieve the design while ensuring that adjacent pieces have corresponding dimensions to provide a consistent stitch edge. Further, the default fabric pieces or default material pieces correlated with the drawing metadata are scaled to match the design sketch (404), and subsequently, a first 3D clothing surface (406) is generated by rendering the assembled clothing on a virtual mannequin using the material information (401) and the cut pieces (402) including, but not limited to, patterns, where the first 3D clothing surface (406) provides perceptual constraints for pattern optimization.

[0027] Continuing with each iteration of the optimization process, the first 3D garment surface (406) and the second 3D garment surface resulting from the original pattern are compared until the deviation between the first 3D garment surface (406) and the second 3D garment surface is reduced to a minimum threshold. In some embodiments, the deviation is measured as the chamfer distance between these two surfaces. All pairs of adjacent patches on the first 3D garment surface (406) are checked, and patch merging (407) is performed for a pair of patches if the curvature across the sewing edge is less than a predetermined threshold and the edges to be merged are small enough to fit within the dimensions of the source fabric. Merging patches on the first 3D garment surface (406) reduces the number of patches and results in fewer cuts and seams, thereby providing improved efficiency during garment manufacturing. To merge patches, the start and end points of the edges are aligned. Two corresponding edges are disposed of to form a single patch. In some embodiments, all curvatures of the seams are sorted in ascending order and the merge process starts with the pair of flattest pieces and proceeds towards more curved pieces.

[0028] Subsequent to patch merge (407), patch shape optimization (408) is performed across the 2D fabric patch pieces or 2D material patch pieces, where the objective is to make the shape of each piece more regular, thereby making the cutting of the fabric or material more efficient, making the pieces fit together more easily, and minimizing fabric or material waste. Each patch is represented as a set of predefined shapes (including straight or curved lines), such as polygons assembled to reflect the curvature of the patch according to, but not limited to, the first 3D garment surface (406). Further, by dividing the patch along predefined shapes, such as but not limited to polygonal boundaries with sharp connections, the curved segments of each patch are identified, and for each adjacent pair of curved segments, the boundary edges are downsampled by gradually reducing the number of vertices along the boundary. Downsampling the boundary edges of the curved segments is stopped when the deviation of the first 3D garment surface (406) reaches a predefined threshold.

[0029] Subsequently, 2D strip packing (409) is performed on the optimized patch shape, where the minimum bounding box is calculated for each piece after packing, followed by calculating the ratio of the empty area for each bounding box, and patches with a ratio exceeding a predetermined threshold are selected. For the selected patches that are divided into smaller patches to achieve maximum space optimization, patch splitting (410) is performed. In some embodiments, the selected patches are split by cutting at each concave edge to eliminate concavity and ensure that all divided pieces are convex. In various embodiments, for the task of strip packing (409), a cuckoo search algorithm based on pairwise clustering is applied. However, it will be recognized that various alternative algorithms may be employed, such as, but not limited to, the bottom-up left-justified algorithm, the next-fit decreasing-height algorithm, Sleator's algorithm, the reverse-fit algorithm, or Steinberg's algorithm. As a result of the steps related to patch merge (407), patch shape optimization (408), strip packing (409), and patch splitting (410), a design pattern output is generated in step (411).

[0030] Furthermore, additional clothing (412) is incrementally increased following patch splitting (410) until a zero-waste configuration or minimum-waste configuration is achieved. The processes of patch merging (407), patch shape optimization (408), strip packing (409), and patch splitting (410) are repeated until a zero-waste replacement or minimum-waste replacement is determined that has the minimum fabric or material consumption, minimum cutting, and maximum 3D surface similarity compared to the second 3D clothing surface resulting from the original pattern. In some embodiments, the processes of patch merging (407), patch shape optimization (408), strip packing (409), and patch splitting (410) may be repeated for multiple pieces of clothing to collectively optimize the production of a set of clothing. For example, if a zero-fabric / material waste design cannot be generated for a single piece of clothing (e.g., a T-shirt), steps (407) through (410) may be repeated for two identical pieces of clothing (or other articles) to be cut from the same fabric. In such embodiments, the cuts between the pieces of clothing are not the subject of the patch merging of step 407. It will be understood that the process described in FIG. 4 prioritizes minimizing design deviation while eliminating as much fabric or material waste as possible within the fabric or material dimensions. In additional embodiments, zero waste is prioritized while allowing the design to be further adapted to fit within the fabric or material dimensions.

[0031] In an alternative embodiment, when the first 2D sewing pattern is in a compact form, the pattern adjustment technique is directly applied to the compact template to retain the zero fabric / material waste property. To implement the pattern adjustment technique, annotations are made on the handles (usually the corners of the cutting pieces) on the pattern template. During pattern adjustment, free movement of the corners is allowed in the design plan, provided there are no self-intersections. The corner positions are optimized such that the design parameters from the input sketch are best reflected in the dimensions from the 2D template. To maintain the feasibility of the manufactured design output, constraints are applied to the dimensions to regularize the design, where the constraints are expressed as a set of linear inequalities on the corner positions.

[0032] Figures 4a, 4b, 4c, and 4d show zero-waste designs across different sizes / body shapes for the same style. For example, a long-sleeved dress in a zero-waste design that is adjusted across S (small) size (shown in FIGS. 4a and 4b), M (medium) size (shown in FIG. 4c), and L (large) size (shown in FIG. 4d) is envisioned. The pattern adjustment technique enables the zero fabric / material waste design to scale across different body or garment measurements (e.g., chest circumference, length of the upper part of the garment), or shapes while maintaining the zero fabric / material waste configuration. This is in contrast to many ready-to-wear garment designs that, while available in multiple standard sizes, cannot be sized to specific measurements or shapes.

[0033] Figures 4e, 4f, and 4g show zero-waste designs across different fabric widths for the same style. Since brands and designers work with multiple suppliers, they often face the problem of inconsistent fabric widths even when using the same materials. Even within the same fabric supplier, there are often variations of 1.27 cm to 5.08 cm (0.5 inches to 2 inches) in width or usable width for each piece of fabric, resulting in fabric waste. The pattern adjustment technique provided in an alternative embodiment of the present invention enables a given zero-waste design to scale across different fabric widths or material dimensions. For example, a zero-fabric-waste hoodie design adjusted to maintain zero fabric waste across three different fabric widths, such as widths of 154.9 cm (61 inches) (shown in FIG. 4e), 162.6 cm (64 inches) (shown in FIG. 4f), and 179.1 cm (70.5 inches) (shown in FIG. 4g), is envisioned. The pattern adjustment technique allows zero-waste designs to scale across different fabric widths, thereby enabling scalability across rolls of fabric with the same width.

[0034] In another embodiment of the present invention, zero-waste design can be achieved using curves as shown in FIG. 4h. The multi-task optimization method disclosed in the present invention can be applied to curves treated as combinations of straight lines. This embodiment is highly adaptable for tailoring and fitting purposes.

[0035] Figures 4i-1 and 4i-2 show a zero fabric waste design that can be achieved by swapping different sections of a given design pattern, changing the size, in order to obtain another design pattern output that can be of different sizes and patterns. In one example, a jacket with dimensions of 152.4 cm x 121.9 cm (60 x 48 inches) is assumed. Using the multi-task optimization method disclosed in the present invention, specific sections such as the cuffs of a given jacket are swapped, rearranged, or strategically sized in different sections to increase the size of the jacket from 152.4 cm x 121.9 cm (60 x 48 inches) to 152.4 cm x 129.5 cm (60 x 51 inches). This enables the use of the same zero waste design style even if the body or clothing measurements are different.

[0036] Figure 5 shows a process for reducing fabric or material waste in material processing according to an embodiment of the present disclosure. In a first approach to waste reduction, an existing zero waste design is used as a template to reduce the computational complexity of patch optimization and 2D strip packing. A parametric model is established to deform the template design to get closest to the target design, while keeping the fabric or material waste zero during such operation. The process (500) includes providing information regarding a plurality of cut pieces (501) and material information (502) to a system (100), and categorizing the target article into a plurality of predefined styles including, but not limited to, shirts, pants, jackets, etc. Further, a zero waste design template (503) is obtained from a template library (504), where the zero waste design template (503) is not limited to a single template and may include multiple related parts of a plurality of templates that can be selected based on the predefined style of the target article.

[0037] As described with reference to FIG. 4, a 3D garment surface (505) and a template 3D garment surface (506) are generated, where the 3D garment surface (505) is generated by rendering the assembled clothing on a virtual mannequin. Subsequently, based on the generated 3D garment surface (505) and the template 3D garment surface (506), the zero-waste design template (503) is rescaled to enable accurate alignment of the generated 3D garment surface (505) and the template 3D garment surface (506). In particular, a scale factor may be assigned to each edge (seam) of the pieces in the zero-waste design template (503). Here, the scale factor for each edge (seam) in the zero-waste design template (503) is provided as the ratio between the respective 3D curves between the template 3D garment surface (506) and the generated 3D garment surface (505). Each edge (seam) in the zero-waste design template (503) has a corresponding 3D curve on the corresponding template 3D garment surface (506). The 3D garment surface (505) and the template 3D garment surface (506) are aligned with each other by 3D registration such as, for example, the iterative closest point algorithm according to an embodiment of the present invention.

[0038] Subsequently, the nearest points on the generated 3D garment surface (505) are retrieved by sampling points along the 3D curve of the zero-waste design template (503). Further, the range of displacement (508) between the generated 3D garment surface (505) and the template 3D garment surface (506) is minimized, where the minimization of the range of displacement can be achieved using a local greedy search algorithm according to an embodiment of the present invention. In some embodiments, the mapping from a certain cut displacement to the length change for all edges in the tech pack or pattern can be determined through the training of a shallow neural network. In some embodiments, the neural network takes the cut displacement in 3D as input and provides the edge length for each piece in the corresponding tech pack or pattern as output. In some embodiments, the displacement is sampled randomly, and training data is generated by calculating the edge length.

[0039] Furthermore, the average edge length between the zero-waste design template (503) and the target article is minimized using closed form optimization (509). Here, the closed form optimization for minimizing the average edge length between the zero-waste design template (503) and the target article retains the design space in the zero-waste domain, thereby guaranteeing a zero fabric / material waste output. In one embodiment, a trust region algorithm is used to minimize the average edge length between the zero-waste design template (503) and the target article using closed form optimization (509). Similar to the case of the method (400), in some embodiments, this process (500) may be repeated for multiple garments to collectively optimize the production of a set of garments.

[0040] <Example 1> Example 1 shows a hoodie with zero fabric waste that uses leftover organic cotton and ribbed fabric as fabrics with a specific orientation. This example demonstrates the uniqueness of the zero-waste design method of fabric / material first. In particular, the method described above with respect to FIG. 4 results in a hoodie with 25% less fabric consumption, a simpler construction than conventional hoodies, a non-basic silhouette (i.e., neither a sack nor a flowing kimono-like), and zero fabric waste.

[0041] The input to the zero-waste method (400) includes fabric information and design information. The fabric information includes an organic cotton fabric with a width of 179.1 cm (70.5 inches) having lines running perpendicular to the width of the fabric. Also, ribbed fabric with a width of 114.3 cm (45 inches) is used for the cuffs and bottom of the hoodie, with more prominent lines running perpendicular to the width of the fabric. The design information includes a hoodie style that includes a design sketch and a tech pack for the hoodie.

[0042] In step (401), fabric information was provided. In step (402), the cut pieces of the target hoodie shown in FIG. 6 were provided. Referring to FIG. 6, the main pieces including the front and rear body pieces, sleeve pieces, hood pieces, and pocket pieces will be apparent. The waste is represented by the filled areas between the pieces. In step (406), a 3D surface is generated. In this example, first assuming an S / M size, an initial measurement tolerance was assigned to each piece based on the desired 3D clothing surface. Although a standard size is assumed in this example, the disclosure of the present invention provides the flexibility to customize clothing, shoes, accessories, and other items to the measurements and sizes requested by the end user. In this example, the body length and pockets have the largest tolerance, and the sleeves have the smallest tolerance. The 3D clothing surface also reveals the dependencies between the pieces. The front and rear body pieces are attached to ribbed pieces, and the sleeves are attached to ribbed cuff pieces. This creates a grouping that gives a combined measurement tolerance that replaces the tolerance of each individual piece. In this way, elongation or expansion of the clothing beyond the size constraints is avoided in later steps. In this example, in step (407), it was determined that patch merging was not possible.

[0043] In step (408), patch shape optimization was carried out. The front and rear body panels were changed to rectangles that span both shoulders and extend vertically up to the waist. The dimensions of the front and rear body panels are 66.04 cm (26 inches) horizontally × 71.12 cm (28 inches) vertically for S / M (for L / XL, 66.04 cm (26 inches) × 78.74 cm (31 inches)). In step (409), packing is carried out. For the front and rear body pieces, the arrangement must be perpendicular to the width of the organic cotton fabric so as to allow the fabric threads to run vertically up and down the body. This complies with industry standards. The arrangement also aligns with the fabric edge and maximizes the remaining space within the fabric width to accommodate the longest remaining piece, the sleeve of the hoodie. For packing the remaining pieces (sleeves, hood pieces, pockets), the wide hood piece is most efficient when placed under the wide body piece, and the sleeves are most efficient when used to fill the remainder of the fabric width.

[0044] In steps (410) and (407), splitting and merging are carried out respectively, and in step (408), iterative shape optimization is carried out. The sleeves are split so as to be optimized within a right trapezoid. The shape optimization also changes the hood shape to a right trapezoid in order to fit the hood and longer sleeves while using less fabric length (i.e., reducing fabric consumption). In step (409), packing is reapplied, whereby the pocket is placed in the lower right corner of the remaining slot. This iterative approach continued to minimize cuts and complex cuts while maintaining zero waste and minimizing fabric consumption while adhering to the target design as much as possible.

[0045] In this example, the resulting design was compared with the target design to ensure a similar silhouette. To achieve this, silhouette comparison was performed in 3D surface rendering. In this example, since it was revealed that the hood was sharp, the hood design was further iterated to add seams to obtain a curved rear part of the hood. Further comparison revealed that the total length of the sleeves and cuffs was too short. Increasing the length of the sleeves would force the height of the pocket to deviate from the allowable measurement parameters, so instead, the height of the cuffs was increased.

[0046] Figures 7 and 8 show exemplary hoodie cuts generated according to an embodiment of the present disclosure. Referring to Figure 7, the resulting fabric cut is shown. This design uses 111.6 cm (1.22 yards), which is 25% less than the input design. In addition to the main components of the garment, the cuffs and the bottom hem of the body were optimized to use ribbed fabric that is 114.3 cm (45 inches) wide. For each hoodie, there are two wrist cuff pieces and one long body hem piece. To create room for the two cuffs, the body piece takes too much width, and the cuffs are wider than the body piece. As a result, the resulting rectangle does not fit inside the fabric piece cut into a rectangle without producing waste. Therefore, in the first iteration, it was determined that zero fabric waste was impossible to achieve. Accordingly, in this example, the fabric usage efficiency was improved by combining multiple units of the body hem and cuffs to be made at once. The resulting cut in Figure 8 shows the final arrangement of the hoodie ribbed fabric with zero waste. For additional garment sizes (L / XL sizes), a new tolerance range with new upper and lower limits was set and the process was repeated.

[0047] <Example 2> In this example, the waste zero - first method (500) of FIG. 5 was adopted. The target jacket is shown in FIG. 9. This design uses the leftover denim at the roll end, which is a small rectangular denim fabric piece remaining at the end of the fabric roll. This roll - end material is usually discarded as it is too small for another production run, but here it is used as an area constraint for constructing the original waste - zero design. According to step (404), a design sketch (see FIG. 9) was provided. The jacket style was decomposed into fabric pieces based on the category of conventional jacket styles, and cut pieces (501) were obtained, including front and back body pieces, collar, rear support, sleeves, cuffs, rear letter piece, and pockets. In step (502), material information was provided, which in this example is the roll - end denim piece, a small rectangle of 152.4 cm (60 inches)×43.18 cm - 45.72 cm (17 - 18 inches). Reference fabric waste - zero jacket templates, or multiple related parts of multiple templates, were obtained from the template library (504). With the initial measurement tolerance for each piece, 3D clothing surfaces (505), (506) of both jacket versions were constructed. According to the template library (504), the largest allowable variation widths are for the body length / width and pocket length, and the smallest are for the sleeve width and collar. The sleeves are attached to the cuff pieces. This creates a grouping that gives a combined measurement tolerance that replaces the tolerance for each individual piece.

[0048] Shape optimization is used such that the front body panel is simplified into two hexagons (or a combination of two trapezoids) that span both shoulders and extend vertically to the waist. The process of shape optimization is repeatedly implemented by applying patch splitting or patch merging, thereby converting the pieces of the collar and the back of the neck support into triangles. Further, packing is performed on the front panel, but here, since the denim fabric is of a small width, the panel can only be packed by arranging it along the width of the denim fabric. The remaining area can be packed with the triangular shape optimized for the collar and the back of the neck support. The back panel is similar to the front panel, but has a smaller lower width limit and no front opening. In packing, this is a mirroring of the previous structure with a second piece of denim fabric, and the remaining fabric is for the collar and the back letter pieces. The back letter pieces are shorter than the remaining fabric, and therefore, it should be carefully noted that the process is readjusted so that two back letter pieces are combined for each of the denim fabrics.

[0049] The last denim piece is used to pack the remaining pieces, most of which are sleeves. The sleeves are packed along the corner edges, leaving enough width to accommodate the pieces of the cuffs and the pockets. Shape optimization is reapplied to change the pieces of the cuffs and the pockets into rectangles that fit into the remaining fabric. The pocket piece is determined to be packed at the ends so as to occupy the remaining fabric as it can have the largest variation range of dimensions. The placement, shape, and potential piece merging are repeated to ensure the minimum cut with the maximum fabric usage efficiency. The resulting design pattern is compared with the target design to guarantee a silhouette similar to the desired jacket style. Further iterations are performed until the style is properly aligned with the desired jacket.

[0050] Figure 10a shows an exemplary jacket cut generated according to an embodiment of the present disclosure. Figure 10b shows the finished product of the jacket whose design sketch was provided in Figure 9.

[0051] <Example 3> Example 3 shows a zero material waste shoe design using multiple fabric and material pieces. In this example, the same fabric-first principle as in the upper clothing example is applied. The different fabric and material pieces required for each shoe are mapped, and then the shape and arrangement of the fabric or material are optimized using shape optimization, splitting or merging, and packing. Shoe production is often done in batches, and opportunities for further zero waste can be ensured by packing multiple shoe pieces together within a fabric or material piece. The same process may be similarly applied to other interior objects such as furniture or automotive seats made of materials including fabric, wood, and / or foam, for example, it will be understood that a zero waste sofa or a fabric zero waste / leather zero waste automotive seat can be created.

[0052] <Example 4> Example 4 shows a clothing item with zero material waste using multiple fabrics. The fabric-first method described herein can also be applied to clothing items using multiple fabrics. To accommodate multiple fabrics, the parts required for each fabric are mapped, and then the design is repeated within each fabric so that zero waste is guaranteed between materials.

[0053] <Example 5> Example 5 shows a garment with zero material waste with specific print alignment. Often, a fabric for clothing has a specific print that must be placed in a certain part of the garment (e.g., a flower print on the front of a dress). This can be achieved by first fixing the placement (not the shape) of a piece of a specific fabric, then optimizing the shape, and packing, splitting, or merging the remaining fabric pieces around that first piece according to the fabric / material first method described herein.

[0054] <Example 6> Example 6 shows multiple zero-material-waste designs that are processed in a batch. By applying the method described herein, multiple styles can be designed at once (e.g., a bag and a shirt can be designed together so that no waste fabric is generated between them). This is particularly appropriate when styles are made on the same manufacturing equipment from the same fabric roll or the same material source. In this example, all styles are broken down into patch pieces, and splitting / and in some cases merging, patch shape optimization, and packing are performed together for all patches.

[0055] In addition to the above examples, it will be understood that various modifications are possible. In various embodiments, the end design does not occupy the entire fabric width. For example, the design may be configured to occupy a certain percentage of the fabric width (e.g., one-half of the fabric width), in which case multiple fabric waste zero designs or parts can be combined across the width of the fabric each time (here, twice). In various embodiments, the input includes a pattern, a tech pack, any 2D sketch, and / or 3D art. In any of these inputs, different patch pieces are separated for further optimization. Although the above examples focus on clothing, it will be understood that the methods described herein are applicable to various textile goods including, but not limited to, clothing, bags, accessories, furniture, and shoes. These methods may also be applied to the material waste zero design of any product made of hard materials.

[0056] The various embodiments described herein use a learning system, or a machine learning model. In some such embodiments, a feature vector is provided to the learning system. Based on the input features, the learning system generates one or more outputs. In some embodiments, the output of the learning system is a feature vector. In some embodiments, the learning system is pre-trained using training data. In some embodiments, the training data is retrospective data. In some embodiments, the retrospective data is stored in a data storage. In some embodiments, the learning system may be further trained through manual curation of previously generated outputs.

[0057] In some embodiments, the learning system includes a support vector machine (SVM). In other embodiments, the learning system includes an artificial neural network. In some embodiments, the learning system is a trained classifier. In some embodiments, the trained classifier is a random decision forest. However, it will be understood that various other classifiers, including linear classifiers, SVMs, or neural networks such as recurrent neural networks (RNNs), are suitable for use according to the present disclosure. Suitable artificial neural networks include, but are not limited to, feedforward neural networks, radial basis function networks, self-organizing maps, learning vector quantization, recurrent neural networks, Hopfield networks, Boltzmann machines, echo state networks, long short-term memory, bidirectional recurrent neural networks, hierarchical recurrent neural networks, probabilistic neural networks, modular neural networks, associative neural networks, deep neural networks, deep belief networks, convolutional neural networks, convolutional deep belief networks, large memory storage and retrieval neural networks, deep Boltzmann machines, deep stacking networks, tensor deep stacking networks, spike and slab restricted Boltzmann machines, composite hierarchical deep models, deep coding networks, multi-layer kernel machines, or deep Q networks.

[0058] FIG. 11 shows a method for generating a zero-waste design pattern according to an embodiment of the present disclosure. The method (1100) includes, in step (1101), receiving an input, which may be a target design including fabric or material dimensions, or a first plurality of cut pieces including, but not limited to, a pattern. In one embodiment of the present invention, the target design input is not limited to cut pieces and also extends to input from a template library or 2D or 3D designs. Subsequently, in step (1102), a first 3D garment surface is rendered from the first plurality of cut pieces from the target design input. Step (1103) includes repeatedly applying merge or split, optimization, and packing to the first plurality of cut pieces including, but not limited to, a pattern to produce a second plurality of cut pieces including, but not limited to, a pattern. As a result of step (1103) in which the steps of applying merge or split, optimization, and packing to the first plurality of cut pieces are performed, in step (1104), a zero-waste design output is obtained. Further, in step (1105), a second 3D garment surface is rendered from the second plurality of cut pieces. In step (1106), the first 3D garment surface and the second 3D garment surface are compared, and when the deviation between the first 3D garment surface and the second 3D garment surface exceeds a predefined threshold, the tasks of applying merge or split, optimization, and packing are repeatedly performed.

[0059] FIG. 12 shows a block diagram of a system for reducing fabric or material waste in material processing. The system (100) includes a computing node (10), and the computing node (10) includes a computer server (12) capable of executing a process for reducing fabric or material waste in material processing. The computer server (12) includes: (i) a system memory (28) which is a computer-readable storage medium composed of one or more program modules (42) for implementing a multi-task optimization method for reducing fabric or material waste in material processing, and the system memory (28) includes computer-readable media in the form of removable memory, non-removable memory, volatile memory, and non-volatile memory; (ii) one or more processing units (16) capable of executing the program modules (42) stored in the system memory (28), and the processing units (16) sequentially execute a multi-task optimization method including steps of patch merge, patch shape optimization, strip packing, and patch splitting, which are repeatedly implemented to improve the packing efficiency of clothes; (iii) a network adapter (20) for enabling wired or wireless communication between components of the computer server (12) via a bus (18), and the computer server (12) communicates with one or more external devices (14) via an input / output (I / O) interface (22). The bus (18) represents any one or more of several types of bus structures including a memory bus or memory controller, a peripheral device bus, an accelerated graphics port, and a processor or local bus, using any of various bus architectures.By way of example and not limitation, such architectures include Industry Standard Architecture (ISA) buses, Micro Channel Architecture (MCA) buses, Enhanced ISA (EISA) buses, Video Electronics Standards Association (VESA) local buses, Peripheral Component Interconnect (PCI) buses, Peripheral Component Interconnect Express (PCIe), and Advanced Microcontroller Bus Architecture (AMBA).

[0060] The computing node (10) is merely an example of a suitable computing node and is not intended to suggest any limitation as to the scope of use or functionality of the embodiments described herein. In any event, the computing node (10) is capable of implementing any of the functionality recited above and / or performing any of them. Examples of well-known computing systems, environments, and / or configurations that may be suitable for use with a computer server (12) include, but are not limited to, personal computer systems, server computer systems, thin clients, thick clients, handheld or laptop devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputer systems, mainframe computer systems, and distributed cloud computing environments including any of the above systems or devices, among others.

[0061] The computer server (12) can be described in the general context of computer system executable instructions, such as program modules executed by a computer system. Generally, program modules can include routines, programs, objects, components, logic, data structures, etc. that perform a particular task or implement a particular abstract data type. The computer server (12) may be practiced in a distributed cloud computing environment where tasks are performed by remote processing devices linked via a communication network. In a distributed cloud computing environment, program modules can be located on computer system storage media including both local and remote memory storage devices. The computer server (12) typically includes a variety of computer system readable media. Such media may be any available media accessible by the computer server (12) and include both volatile and nonvolatile media, removable and non-removable media.

[0062] The computer server (12) can further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, a storage system (34) may be provided for reading from and writing to a non-removable non-volatile magnetic medium (not shown, commonly referred to as a “hard drive”). Although not shown, a magnetic disk drive for reading from and writing to a removable non-volatile magnetic disk (e.g., a “floppy (registered trademark) disk”), and an optical disk drive for reading from or writing to a removable non-volatile optical disk such as a CD-ROM (compact disc read-only memory), a DVD-ROM (digital versatile disk-read-only memory), or other optical media may be provided. In such examples, each may be connected to the bus (18) by one or more data medium interfaces. As further represented and described below, the system memory (28) can include at least one program product having a set of program modules (e.g., at least one) configured to perform the functions of the embodiments of the present disclosure.

[0063] The system memory (28) can include a computer system readable medium in the form of volatile memory such as random access memory (RAM) (30) and / or cache memory (32). By way of example and not limitation, a program / utility (40) having a set (at least one) of program modules (42) may be stored in the system memory (28), as well as an operating system, one or more application programs, other program modules, and program data. Each of the operating system, one or more application programs, other program modules, and program data, or some combination thereof, can include an implementation of a networking environment. The program modules (42) generally implement the functions and / or methodologies of the examples as described herein.

[0064] The computer server (12) can also communicate with a plurality of external devices (14) such as a keyboard, a pointing device, a display (24), etc., a plurality of devices that enable a user to interact with the computer server (12), and / or any device (e.g., a network card, a modem, etc.) that enables the computer server (12) to communicate with one or more other computing devices. Such communication may be performed via an input / output (I / O) interface (22). Still further, the computer server (12) can communicate with one or more networks such as a local area network (LAN), a general wide area network (WAN), and / or a public network (e.g., the Internet) via a network adapter (20). Although not shown, it should be understood that other hardware and / or software components may be used in conjunction with the computer server (12). For example, but not limited to, microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data archive storage systems, etc.

[0065] The present disclosure may be embodied as a system, method, and / or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions for causing a processor to perform aspects of the present disclosure. The computer-readable storage medium may be a tangible device that can store and retain instructions for use by an instruction execution device. The computer-readable storage medium may be, for example, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of the computer-readable storage medium includes a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a CD-ROM, a DVD, a memory stick, a floppy disk, a mechanically encoded device such as a punch card or raised structure in a groove having instructions recorded thereon, and any suitable combination of the foregoing. The computer-readable storage medium should not be construed as being a transitory signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., an optical pulse passing through an optical fiber cable), or an electrical signal transmitted through a wire.

[0066] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to respective computing / processing devices, or to an external computer or external storage device via a network such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and transfers the computer-readable program instructions for storage in a computer-readable storage medium within each respective computing / processing device.

[0067] Computer-readable program instructions for carrying out the operations of this disclosure may be source code or object code written in any combination of one or more programming languages, including assembly instructions, instruction-set-architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state-setting data, or object-oriented programming languages such as Smalltalk, C++, and the like, and conventional procedural programming languages such as the "C" programming language or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection to the external computer may be made (e.g., via the Internet using an Internet service provider). In some embodiments, for example, an electronic circuit including a programmable logic circuit, a field-programmable gate array (FPGA), or a programmable logic array (PLA) may execute the computer-readable program instructions by utilizing the state information of the computer-readable program instructions for personalizing the electronic circuit to implement aspects of this disclosure.

[0068] Aspects of the present disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions. These computer-readable program instructions are provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions / acts specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions may also be stored in a computer-readable storage medium that can direct a computer, a programmable data processing apparatus, and / or other devices to function in a particular manner, such that the computer-readable storage medium containing the instructions comprises an article of manufacture including instructions which implement the aspects of the function / act specified in one or more blocks of the flowchart and / or block diagram.

[0069] The computer-readable program instructions may also be loaded onto a computer, other programmable apparatus, or other device to produce a computer-implemented process, such that the instructions which execute on the computer, other programmable apparatus, or other device implement the functions / acts specified in one or more blocks of the flowchart and / or block diagram by causing a series of operational steps to be performed on the computer, other programmable apparatus, or other device.

[0070] The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagram can represent a module, segment, or portion of one or more executable instructions for implementing the specified logical function. In some alternative implementations, the functions noted in the blocks may be performed out of the order illustrated. For example, depending on the functionality involved, two blocks shown in succession may actually be executed substantially simultaneously, or the blocks may sometimes be executed in the reverse order. It will also be noted that each block of the block diagrams and / or flowchart diagrams, as well as combinations of blocks in the block diagrams and / or flowchart diagrams, may be implemented by a special purpose hardware-based system that performs the specified function or operation, or a combination of special purpose hardware and computer instructions.

[0071] The descriptions of the various embodiments of the present disclosure have been presented for purposes of illustration, but are not intended to be exhaustive or limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein was chosen to best explain the principles of the embodiments, the practical application, or technical improvements in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A multi - task optimization method for generating a zero - waste design pattern and reducing fabric or material waste, the method (400) comprising: a) providing to a system (100) one or more inputs required for material processing, wherein the provided input is directly taken from material information (401), a pattern or tech - pack (403), or a cut - piece (402) obtained from a design sketch (404), and / or includes metadata indicating additional characteristics required for the material processing; b) retrieving from a template library (405) a combination of templates or template parts based on the provided input, wherein the template houses a set of default fabric pieces or default material pieces correlated with drawing metadata; c) scaling the default fabric piece or default material piece correlated with the drawing metadata to fit the design sketch (404); d) generating a first 3D clothing surface (406) by rendering the assembled clothing on a virtual mannequin using the material information (401) and the cut - piece (402), wherein the first 3D clothing surface (406) provides perceptual constraints for pattern optimization; e) in each iteration of the optimization process, comparing the first 3D clothing surface (406) with a second 3D clothing surface resulting from an original pattern or 2D / 3D sketch until the deviation between the first 3D clothing surface (406) and the second 3D clothing surface is reduced to a minimum threshold; f) checking pairs of adjacent patches on the first 3D clothing surface (406), a. where the curvature across the sewing edge is less than a predetermined threshold and b. the edges to be merged are small enough to fit within the dimensions of the source fabric or source material, performing patch merge (407) for the pair of patches. g) Performing patch shape optimization (408) on a 2D fabric patch piece or 2D material patch piece, each patch being represented as a set of predefined shapes assembled to reflect the curvature of the patch according to the first 3D clothing surface (406); h) Identifying curved segments of each patch by dividing the patch along the boundaries of the predefined shapes having sharp connections, the boundary edges being downsampled by reducing the number of vertices along the boundary for each adjacent pair of curved segments; i) Performing 2D strip packing (409) on the optimized patch shapes, a minimum bounding box being calculated for each piece after packing; j) Calculating the ratio of the empty area for each bounding box and selecting patches for which the ratio exceeds a predetermined threshold, patch splitting (410) being performed on the selected patches that are divided into smaller patches to achieve maximum space optimization; comprising method (400). **Claim 2** The method (400) according to claim 1, wherein the material information (401) includes material type, weight, dimensions, description, texture direction or orientation based on print, and length of the print. **Claim 3** Merging the first 3D clothing surface (406) patches reduces the number of patches and the number of cuts and seams, thereby resulting in improved efficiency during clothing manufacture, the method (400) according to claim 1. **Claim 4** Downsampling the boundary edges of the curved segments is stopped when the deviation of the first 3D clothing surface (406) reaches a predefined threshold, the method (400) according to claim 1. **Claim 5** The patch merging (407), the patch shape optimization (408), the strip packing (409), and the patch splitting (410) are repeatedly performed until zero waste substitution is determined, thereby resulting in minimum fabric or material consumption, minimum cuts, and maximum 3D surface similarity compared to the second 3D clothing surface resulting from the original pattern, the method (400) according to claim 1. **Claim 6** The method (400) according to claim 1, wherein an additional garment (411) is incrementally increased following the patch division (410) until a zero-waste configuration is obtained.

7. A process (500) for reducing fabric or material waste while prioritizing waste reduction over design retention, a. providing information regarding one or more cut pieces (501), 2D / 3D design concepts, and material information (502) to the system (100), and categorizing the target article into one or more predefined styles; and b. obtaining a zero-waste design template (503) from a template library (504), wherein the zero-waste design template (503) is selected based on the predefined style of the target article or directly from the design sketch and the material information; and c. generating a 3D garment surface (505) and a template 3D garment surface (506), wherein the 3D garment surface (505) is generated by rendering the assembled garments on a virtual mannequin; and d. rescaling the zero-waste design template (503) based on the generated 3D garment surface (505) and the template 3D garment surface (506) to enable accurate alignment of the generated 3D garment surface (505) and the template 3D garment surface (506); and e. sampling points along the 3D curve of the zero-waste design template (503) to extract the nearest points on the generated 3D garment surface (505); and f. minimizing the range of displacement (508) between the generated 3D garment surface (505) and the template 3D garment surface (506); and g. minimizing the average edge length between the zero-waste design template (503) and the target article using closed-form optimization (509); and h. optimizing the production of a set of garments collectively by repeating the process (500) for multiple garments comprising the method (400) according to claim 1.

8. The method (400) according to claim 1, wherein the scale factor for each edge in the waste zero design template (503) is provided as a ratio between each 3D curve of the template 3D clothing surface (506) and the generated 3D clothing surface (505).

9. The method (400) according to claim 1, wherein the range of the displacement between the generated 3D clothing surface (505) and the template 3D clothing surface (506) is minimized using a local greedy search algorithm.

10. The method (400) according to claim 1, wherein the closed-form optimization for minimizing the average edge length between the waste zero design template (503) and the target article retains the design space in the waste zero domain, thereby guaranteeing a waste zero output.

11. A method (1100) for generating a waste zero design pattern and reducing fabric or material waste includes a. receiving a target design input including a first plurality of cut pieces; b. rendering a first 3D clothing surface from the first plurality of cut pieces from the target design input; c. repeatedly applying merge or split, optimization, and packing to the first plurality of cut pieces to produce a second plurality of cut pieces; d. obtaining a waste zero design pattern output; e. rendering a second 3D clothing surface from the second plurality of cut pieces; and f. comparing the first 3D clothing surface and the second 3D clothing surface, and repeatedly performing the task of applying merge or split, optimization, and packing when the deviation between the first 3D clothing surface and the second 3D clothing surface exceeds a predefined threshold. The method (400) according to claim 1.

12. A system for reducing fabric or material waste in material processing, the system (100) comprising a. a computing node (10) comprising a computer server (12) capable of executing a process for reducing the fabric or material waste in material processing and the computer server (12) being i. A system memory (28), which is a computer-readable storage medium composed of one or more program modules (42) for implementing a multi-task optimization method for reducing the fabric or material waste in material processing; ii. One or more processing units (16) capable of executing the program modules (42) stored in the system memory (28), the processing unit (16) sequentially executing the multi-task optimization method including steps of patch merging, patch shape optimization, strip packing, and patch splitting, which are repeatedly performed to improve the packing efficiency of clothes; iii. A network adapter (20) for enabling wired or wireless communication between components of the computer server (12) via a bus (18). A system (100) comprising the above.

13. The system (100) according to claim 12, wherein the computer server (12) communicates with one or more external devices (14) via an input / output (I / O) interface (22).

Images