AI Paper Atlas

Research map / 3D vision / Face and shape

Learning Implicit Templates for Point-Based Clothed Human Modeling

Siyou Lin, Hongwen Zhang, Zerong Zheng, Ruizhi Shao, Yebin Liu

ECCV 2022 · 3D vision › Face and shape · approach: Clothed human modeling · 46 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use point-based representation (FITE, POP) instead of implicit methodsFITE · CAPE and ReSynth interpolationd_cham (x1e5): SNARF 2.426 -> FITE 1.355 (Felice 004 long dress)Improved
learn implicit outfit template instead of fixed minimal body templateFITE vs POP · ReSynth Carla 004 long pantsd_cham (x1e5): 0.485 -> 0.299Improved
learn implicit outfit template instead of fixed minimal body templateFITE vs POP · ReSynth Felice 004 long dressd_cham (x1e5): 1.718 -> 1.355Improved
learn implicit outfit template instead of fixed minimal body templateFITE vs POP · CAPE 03375 blazerlongd_cham (x1e5): 0.252 -> 0.274Hurt
learn implicit outfit template instead of fixed minimal body templateFITE vs POP · ReSynth Christine 027 short dressd_cham (x1e5): 0.421 -> 0.455Hurt
cross-outfit training (FITE vs outfit-specific FITE-OS) (ablation)FITE · ReSynth Felice 004 long dressd_cham (x1e5): 1.805 -> 1.355Improved
cross-outfit training (FITE vs outfit-specific FITE-OS) (ablation)FITE · CAPE tight outfitsd_cham (x1e5): 0.041 -> 0.042 (00096), 0.074 -> 0.076 (00215)No clear effect
FITE instead of POP, user preference on point cloudsFITE · Extrapolation user studyvote share (%): 24.58 -> 75.42Improved
FITE instead of POP, user preference on meshesFITE · Extrapolation user studyvote share (%): 40.63 -> 59.37Improved
fix diffused skinning weights instead of jointly learning skinning field (ablation)FITE stage one · template trainingqualitativeImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy