Papers › 3D Multi-bodies: Fitting Sets of Plausible 3D Human Models to Ambiguous Image Data

3D Multi-bodies: Fitting Sets of Plausible 3D Human Models to Ambiguous Image Data

2 Nov 2020NeurIPS 2020 12arXiv:2011.00980archive 2025-07-28

Benjamin Biggs, Sébastien Ehrhadt, Hanbyul Joo, Benjamin Graham, Andrea Vedaldi, David Novotny

We consider the problem of obtaining dense 3D reconstructions of humans from single and partially occluded views. In such cases, the visual evidence is usually insufficient to identify a 3D reconstruction uniquely, so we aim at recovering several plausible reconstructions compatible with the input data. We suggest that ambiguities can be modelled more effectively by parametrizing the possible body shapes and poses via a suitable 3D model, such as SMPL for humans. We propose to learn a multi-hypothesis neural network regressor using a best-of-M loss, where each of the M hypotheses is constrained to lie on a manifold of plausible human poses by means of a generative model. We show that our method outperforms alternative approaches in ambiguous pose recovery on standard benchmarks for 3D humans, and in heavily occluded versions of these benchmarks.

PaperPDFConference PDF

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

3D ReconstructionMulti-Hypotheses 3D Human Pose Estimation

Datasets

Introduced by this paper, per the archive.

AH36M

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Multi-Hypotheses 3D Human Pose Estimation AH36M 3D Multi-bodies Best-Hypothesis MPJPE (n = 25) 90.0 #3 of 10 Archive leaderboard report
Multi-Hypotheses 3D Human Pose Estimation AH36M 3D Multi-bodies Best-Hypothesis PMPJPE (n = 25) 64.2 #3 of 10 Archive leaderboard report
Multi-Hypotheses 3D Human Pose Estimation AH36M 3D Multi-bodies H36M PMPJPE (n = 1) 41.6 #3 of 10 Archive leaderboard report
Multi-Hypotheses 3D Human Pose Estimation AH36M 3D Multi-bodies H36M PMPJPE (n = 25) 42.2 #3 of 10 Archive leaderboard report
Multi-Hypotheses 3D Human Pose Estimation AH36M 3D Multi-bodies Most-Likely Hypothesis PMPJPE (n = 1) 67.8 #3 of 10 Archive leaderboard report
Multi-Hypotheses 3D Human Pose Estimation AH36M SMPL-CVAE Best-Hypothesis MPJPE (n = 25) 109.7 #6 of 10 Archive leaderboard report
Multi-Hypotheses 3D Human Pose Estimation AH36M SMPL-CVAE Best-Hypothesis PMPJPE (n = 25) 75.1 #6 of 10 Archive leaderboard report
Multi-Hypotheses 3D Human Pose Estimation AH36M SMPL-CVAE H36M PMPJPE (n = 1) 46.7 #6 of 10 Archive leaderboard report
Multi-Hypotheses 3D Human Pose Estimation AH36M SMPL-CVAE H36M PMPJPE (n = 25) 46.2 #6 of 10 Archive leaderboard report
Multi-Hypotheses 3D Human Pose Estimation AH36M SMPL-CVAE Most-Likely Hypothesis PMPJPE (n = 1) 76.5 #6 of 10 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections