AI Paper Atlas

Research map / 3D vision / Human pose estimation

Humans in 4D: Reconstructing and Tracking Humans with Transformers

Shubham Goel, Georgios Pavlakos, Jathushan Rajasegaran, Angjoo Kanazawa, Jitendra Malik

ICCV 2023 · 3D vision › Human pose estimation · approach: Human mesh recovery from images · 344 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
replace ResNet-50 backbone with ViT-H (ablation)HMR · LSP-ExtendedPCK@0.05: 0.35 -> 0.48Improved
add AVA data to training set (ablation)HMR 2.0 · 3DPW testMPJPE (mm): 84.1 -> 80.2Improved
train from random ViT initialization instead of MAE plus 2D keypoint pretraining (ablation)HMR 2.0 · 3DPW testMPJPE (mm): 81.3 -> 98.9Hurt
add 2D keypoint pretraining after MAE (ablation)HMR 2.0 · LSP-ExtendedPCK@0.05: 0.52 -> 0.53Improved
use transformer decoder head instead of MLP IEF head (ablation)HMR 2.0 · LSP-ExtendedPCK@0.05: 0.48 -> 0.53Improved
train longer on more data (HMR 2.0b vs 2.0a) (ablation)HMR 2.0 · 3DPW testMPJPE (mm): 70.0 -> 81.3Mixed
use HMR 2.0 as pose engine in PHALP' trackingPHALP' · PoseTrack valID switches: 623 -> 456Improved
use full 4DHumans instead of PHALPtracking · PoseTrack valID switches: 541 -> 421Improved
use ViTDet detector instead of Mask R-CNN in 4DHumans4DHumans · PoseTrack valHOTA: 54.3 -> 57.8Improved
use HMR 2.0 poses for pose-based action recognitionLART pose-only · AVA v2.2mAP: 19.6 -> 22.3Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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