AI Paper Atlas

Research map / 3D vision / Human pose estimation

Slim DensePose: Thrifty Learning From Sparse Annotations and Motion Cues

Natalia Neverova, James Thewlis, Riza Alp Guler, Iasonas Kokkinos, Andrea Vedaldi

CVPR 2019 · 3D vision › Human pose estimation · approach: Temporal video pose estimation · 32 citations (OpenAlex)

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 sparse UV annotations on more images rather than complete annotations on a small image subset (ablation)Hourglass DensePose model · DensePose-COCO5 cm: 36.27 -> 47.68Improved
Add part labels across images while sparsifying UV labels (ablation)Hourglass DensePose model · DensePose-COCO5 cm: 37.65 -> 40.80Improved
Add keypoint annotations to 1% UV supervision (ablation)Hourglass DensePose model · DensePose-COCO5 cm: 37.65 -> 39.17Improved
Use segmentation-only supervision instead of UV correspondence labels (ablation)Hourglass DensePose model · DensePose-COCO5 cm: 55.78 -> 3.53Hurt
Pretrain on COCO before fine-tuning on DensePose-TrackHourglass DensePose model · DensePose-Track5 cm: 21.06 -> 45.57Improved
Fine-tune on the union of COCO and Track after COCO pretrainingHourglass DensePose model · DensePose-Track5 cm: 45.57 -> 46.04Improved
Add ground-truth propagation using optical flow (ablation)Hourglass DensePose model · DensePose-Track20 cm: 59.54 -> 75.95Improved
Add equivariance using optical flow (ablation)Hourglass DensePose model · DensePose-Track20 cm: 59.54 -> 62.22Improved
Combine optical-flow GT propagation and equivariance (ablation)Hourglass DensePose model · DensePose-Track20 cm: 59.54 -> 77.20Improved
Use real optical flow instead of synthetic TPS motion (ablation)Hourglass DensePose model with GT propagation and equivariance · DensePose-Track20 cm: 62.71 -> 77.20Improved
Add optical-flow GT propagation and equivariance after COCO pretraining (ablation)Hourglass DensePose model · DensePose-Track20 cm: 83.71 -> 87.56Improved
Train jointly on COCO and Track without staged fine-tuning (ablation)Hourglass DensePose model · DensePose-Track5 cm: 44.89 -> 41.76Hurt

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