AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Unsupervised Magnification of Posture Deviations Across Subjects

Michael Dorkenwald, Uta Buchler, Bjorn Ommer

CVPR 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Dense features and foundation priors · 13 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
Magnify posture deviations across subjects using the proposed method instead of using original imagesHG2DB impairment classification · HG2DBaccuracy: 56.5 ± 4.5 -> 67.7 ± 3.5Improved
Use the proposed posture magnification instead of Oh et al. [43] magnificationsHG2DB impairment classification · HG2DBaccuracy: 57.2 ± 4.4 -> 67.7 ± 3.5Improved
Add the disentanglement loss Ldis (ablation)Posture magnification model · HG2DBvisual generation qualityImproved
Add the magnification loss Lmag (ablation)Posture magnification model · HG2DBvisual generation qualityImproved
Remove Lmag (ablation)Posture magnification model · HG2DBFIDHurt
Remove both Ldis and Lmag (ablation)Posture magnification model · HG2DBvisual generation qualityHurt

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