AI Paper Atlas

Research map / 3D vision / Human pose estimation

3D Human Pose Estimation via Non-Causal Retentive Networks

Kaili Zheng, Feixiang Lu, Yihao Lv, Liangjun Zhang, Chenyi Guo, Ji Wu

ECCV 2024 · 3D vision › Human pose estimation · approach: Monocular 2D-to-3D pose lifting · 5 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
train with a large chunk (243) and infer with a small chunk (27) instead of training at 27 (ablation)NC-RetNet · Human3.6M (CPN), test T=27MPJPE (mm): 43.7 -> 42.1Improved
train with chunk 243 rather than 81, test at T=81 (ablation)NC-RetNet · Human3.6M (CPN), test T=81MPJPE (mm): 41.9 -> 40.9Improved
add cross-chunk state to the RetNet model (ablation)NC-RetNet · Human3.6M (CPN), train T=243, test T=27MPJPE (mm): 54.2 -> 42.1Improved
add cross-chunk state, test at the same size as training (T=243) (ablation)NC-RetNet · Human3.6M (CPN), test T=243MPJPE (mm): 42.5 -> 40.4Improved
replace MixSTE temporal encoder with xPos (no state) versus NC-RetNet (ablation)MixSTE · Human3.6M (CPN), train T=243, test T=27MPJPE (mm): 48.8 vs 42.1Hurt
apply NC-RetNet instead of STCFormer at small test chunk sizeHuman3.6M CPN · Human3.6M (CPN), T=27 vs 81MPJPE (mm): 42.1 (Ours T=27) vs 42.0 (STCFormer T=81)Improved
NC-RetNet versus STCFormer at large test chunk (T=243) on PA-MPJPEHuman3.6M CPN · Human3.6M (CPN), T=243PA-MPJPE (mm): 32.5 vs 31.8 (STCFormer)Hurt
use NC-RetNet with small chunk size on MPI-INF-3DHPMPI-INF-3DHP · MPI-INF-3DHP, T=9MPJPE (mm): 28.2 -> 24.1Improved
use NC-RetNet with 2D ground truth keypoints at T=81Human3.6M GT · Human3.6M (2D ground truth input), T=81MPJPE (mm): 25.7 -> 22.4Improved
replace MixSTE temporal encoder with NC-RetNet, compute costMixSTE · Human3.6M (CPN), T=243FLOPs per frame (M): 572 -> 430Improved

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