Human Shape and Clothing Estimation
Aayush Gupta, Aditya Gulati, Himanshu, Lakshya LNU
TL;DR
SMPL-based 3D body modeling with shape $\\beta$ and pose $\\theta$ underpins modern human shape estimation, while virtual try-on, landmark detection, and attribute recognition enable realistic and personalized fashion experiences. The survey covers four intertwined tasks: human shape estimation, fashion generation, landmark detection, and attribute recognition, detailing foundational methods (e.g., HMR, SPIN) and garment synthesis approaches (TPS-based, dense-flow, and 3D cloth deformations) as well as attention- and transformer-based landmark/attribute techniques, including DETR-based segmentation. Key contributions include clarifying SMPL's central role, comparing garment warping strategies, and tracing the shift from pose-dependent pipelines to attention and transformer-based frameworks. The work provides a pragmatic roadmap for building scalable, real-time, and application-ready systems in fashion and immersive media, while noting ongoing challenges such as dataset biases and computational constraints.
Abstract
Human shape and clothing estimation has gained significant prominence in various domains, including online shopping, fashion retail, augmented reality (AR), virtual reality (VR), and gaming. The visual representation of human shape and clothing has become a focal point for computer vision researchers in recent years. This paper presents a comprehensive survey of the major works in the field, focusing on four key aspects: human shape estimation, fashion generation, landmark detection, and attribute recognition. For each of these tasks, the survey paper examines recent advancements, discusses their strengths and limitations, and qualitative differences in approaches and outcomes. By exploring the latest developments in human shape and clothing estimation, this survey aims to provide a comprehensive understanding of the field and inspire future research in this rapidly evolving domain.