English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Dressed people reconstruction from images is a popular task with promisingapplications in the creative media and game industry. However, most existingmethods reconstruct the human body and garments as a whole with the supervisionof 3D models, which hinders the downstream interaction tasks and requireshard-to-obtain data. To address these issues, we propose an unsupervisedseparated 3D garments and human reconstruction model (USR), which reconstructsthe human body and authentic textured clothes in layers without 3D models. Morespecifically, our method proposes a generalized surface-aware neural radiancefield to learn the mapping between sparse multi-view images and geometries ofthe dressed people. Based on the full geometry, we introduce a Semantic andConfidence Guided Separation strategy (SCGS) to detect, segment, andreconstruct the clothes layer, leveraging the consistency between 2D semanticand 3D geometry. Moreover, we propose a Geometry Fine-tune Module to smoothedges. Extensive experiments on our dataset show that comparing withstate-of-the-art methods, USR achieves improvements on both geometry andappearance reconstruction while supporting generalizing to unseen people inreal time. Besides, we also introduce SMPL-D model to show the benefit of theseparated modeling of clothes and the human body that allows swapping clothesand virtual try-on.

USR: Unsupervised Separated 3D Garment and Human Reconstruction via Geometry and Semantic Consistency