GAN-Based Garment Generation Using Sewing Pattern Images

NeuralTailor: Reconstructing Sewing Pattern Structures from 3D Point Clouds of Garments

“…We believe that reconstructing disentangled garment representations will eventually lead to better quality, control, and generalization. Shen et al [2020] demonstrate a garment model generator conditioned on sewing patterns with capabilities to generalize to novel designs. In our work, we show that using sewing pattern as a natural structured representation of design when inferring it from raw inputs allows not only for generalization to unseen garment examples as in [Ma et al 2021], but unseen garment types.…”

The Power of Points for Modeling Humans in Clothing

“…To overcome the limitations brought by the fixed topology of meshes, recent work opts for other representations that can unify different clothing categories and types. Shen et al [57] represent garments using 2D sewing pattern images that are applicable to arbitrary clothing categories. However, the final 3D garment shape is represented with a single manifold mesh that is not sufficiently expressive to represent the complexity and variety of real-world clothing.…”

Deep Detail Enhancement for Any Garment

“…While this is a very efficient representation, it only works well for tight clothes such as t‐shirt or pants that are sufficiently close to the body surface. Shen et al [SLL20] use a GAN‐based method to generate garment models by transferring the sewing patterns of different garments to the UV space of the underlying garment. Gundogdu et al [GCS*19] present GarNet which learns features of garment deformation as a function of body pose and shape at different levels (point‐wise, patch‐wise, and global features) to reconstruct the detailed draping output.…”