AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Self-supervised Secondary Landmark Detection via 3D Representation Learning

Praneet Bala, Jan Zimmermann, Hyun Soo Park, Benjamin Y. Hayden

IJCV 2023 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Invariance-based self-supervised learning · 11 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 multiview contrastive loss L_U^c (ablation)L_L + L_U^g · Human3.6MPCKh@0.5 mean: 0.43 -> 0.57Improved
add multiview contrastive loss L_U^c (ablation)L_L + L_U^g · DeepFly3DPCKh@0.5 mean: 0.42 -> 0.79Improved
add multiview contrastive loss L_U^c (ablation)L_L + L_U^g · OpenMonkeyPosePCKh@0.5 mean: 0.74 -> 0.75No clear effect
add 3D shared-representation self-supervision L_U^g (ablation)supervised L_L · Human3.6MPCKh@0.5 mean: 0.12 -> 0.43Improved
add 3D shared-representation self-supervision L_U^g (ablation)supervised L_L · DeepFly3DPCKh@0.5 mean: 0.31 -> 0.42Improved
add 3D shared-representation self-supervision L_U^g (ablation)supervised L_L · OpenMonkeyPosePCKh@0.5 mean: 0.72 -> 0.74Improved
add triangulation reprojection self-supervision L_U^t (ablation)supervised L_L · Human3.6MPCKh@0.5 mean: 0.12 -> 0.09Hurt
add triangulation reprojection self-supervision L_U^t (ablation)supervised L_L · DeepFly3DPCKh@0.5 mean: 0.31 -> 0.28Hurt
use 3D instead of 2D primary-landmark representation for predicting secondary landmarks (ablation)linear subspace analysis · OpenMonkeyPosesecondary landmark reconstruction error: 30.2-60.6% error reductionImproved
use semi-supervised multiview training instead of supervised only (2k labeled) (ablation)Ours · OpenMonkeyPose (2k labeled)PCKh secondary landmarks: 0.567 -> 0.413Hurt
use semi-supervised multiview training instead of supervised only (14k labeled) (ablation)Ours · OpenMonkeyPose (14k labeled)PCKh secondary landmarks: 0.722 -> 0.753Improved
remove wrist primary landmarks from the input configuration (ablation)shared-representation predictor · OpenMonkeyPoseelbow reconstruction errorHurt

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