AI Paper Atlas

Research map / 3D vision / Face and shape

AutoAvatar: Autoregressive Neural Fields for Dynamic Avatar Modeling

Ziqian Bai, Timur Bagautdinov, Javier Romero, Michael Zollhöfer, Ping Tan, Shunsuke Saito

ECCV 2022 · 3D vision › Face and shape · approach: Pixel-aligned implicit human and hand reconstruction · 3 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
add autoregressive shape history (H_t and dH) to pose-only input (ablation)Pose+dPose variant · DFaust interpolation, rollout 1Scan-to-prediction distance (mm): 3.852 -> 1.569Improved
add autoregressive shape history (ablation)Pose+dPose variant · DFaust extrapolation, rollout 1Scan-to-prediction distance (mm): 3.984 -> 1.150Improved
add autoregressive shape history (ablation)Pose+dPose variant · DFaust interpolation, rollout 30Scan-to-prediction distance (mm): 4.156 -> 5.255Hurt
add dPose (pose differences) to pose-only input (ablation)Pose · DFaust interpolation, rollout 1Scan-to-prediction distance (mm): 4.218 -> 3.852Improved
replace single-frame pose input with TCN over 16 poses (ablation)Pose · DFaust interpolation, rollout 1Scan-to-prediction distance (mm): 4.218 -> 4.068Improved
encode shape with signed heights on articulated observer points instead of localized latent embeddings (ablation)L-embed · DFaust interpolation, rollout 1Scan-to-prediction distance (mm): 1.784 -> 1.569Improved
encode shape with signed heights instead of localized latent embeddings (ablation)L-embed · DFaust extrapolation, rollout 30Scan-to-prediction distance (mm): 5.192 -> 4.526Improved
encode shape with signed heights instead of global latent embedding (ablation)G-embed · DFaust interpolation, rollout 1Scan-to-prediction distance (mm): 2.932 -> 1.569Improved
encode shape with signed heights instead of global latent embedding (ablation)G-embed · DFaust interpolation, rollout 30Scan-to-prediction distance (mm): 3.793 -> 5.255Hurt
use AutoAvatar instead of SNARFSNARF · DFaust interpolation, rollout 1Scan-to-prediction distance (mm): 7.428 -> 1.569Improved
use refined SMPL poses instead of AMASS posesSNARF · DFaust subject 50002 interpolation, rollout 1Scan-to-prediction distance (mm): 7.898 -> 3.982Improved
use refined SMPL poses instead of AMASS posesAutoAvatar · DFaust subject 50002 interpolation, rollout 1Scan-to-prediction distance (mm): 1.731 -> 1.417Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 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