AI Paper Atlas

Research map / 3D vision / Face and shape

An Adversarial Neuro-Tensorial Approach for Learning Disentangled Representations

Mengjiao Wang, Zhixin Shu, Shiyang Cheng, Yannis Panagakis, Dimitris Samaras, Stefanos Zafeiriou

IJCV 2019 · 3D vision › Face and shape · approach: 3D morphable face model regression · 25 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 proposed tensor latent model instead of prior method 2 for normalsdisentangling network · Photoface normal estimationmean angular error (degrees): 30.09 -> 28.67Improved
use proposed model instead of prior method 2disentangling network · Photoface normal estimationfraction within 35 degrees: 84.6 -> 89.1Improved
use proposed model instead of prior method 2disentangling network · Photoface normal estimationfraction within 40 degrees: 98.1 -> 96.3Hurt
use proposed z_id instead of prior latent C for identitydisentangling network · Multi-PIE (frontal only)identity accuracy: 19.18 -> 99.33Improved
use proposed z_exp instead of prior latent E for expressiondisentangling network · Multi-PIE (frontal only)expression accuracy: 35.49 -> 78.92Improved
use proposed z_ill instead of prior latent L for illuminationdisentangling network · Multi-PIE (frontal only)illumination accuracy: 48.85 -> 64.11Improved
use proposed disentangled Z_exp and Z_p instead of adversarial auto-encoder latent Z_0disentangling network · Multi-PIE t-SNEcluster qualityImproved

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