Timezone: »
We consider the problem of obtaining dense 3D reconstructions of deformable objects 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 modeled 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.
Author Information
Benjamin Biggs (University of Cambridge)
David Novotny (Facebook AI Research)
Sebastien Ehrhardt (University of Oxford)
Hanbyul Joo (FAIR)
Ben Graham (Facebook Research)
Andrea Vedaldi (University of Oxford / Facebook AI Research)
Related Events (a corresponding poster, oral, or spotlight)
-
2020 Poster: 3D Multi-bodies: Fitting Sets of Plausible 3D Human Models to Ambiguous Image Data »
Tue. Dec 8th 05:00 -- 07:00 PM Room Poster Session 1 #407
More from the Same Authors
-
2021 : PASS: An ImageNet replacement for self-supervised pretraining without humans »
Yuki Asano · Christian Rupprecht · Andrew Zisserman · Andrea Vedaldi -
2021 : PASS: An ImageNet replacement for self-supervised pretraining without humans »
Yuki Asano · Christian Rupprecht · Andrew Zisserman · Andrea Vedaldi -
2021 Poster: Keeping Your Eye on the Ball: Trajectory Attention in Video Transformers »
Mandela Patrick · Dylan Campbell · Yuki Asano · Ishan Misra · Florian Metze · Christoph Feichtenhofer · Andrea Vedaldi · João Henriques -
2021 Oral: Keeping Your Eye on the Ball: Trajectory Attention in Video Transformers »
Mandela Patrick · Dylan Campbell · Yuki Asano · Ishan Misra · Florian Metze · Christoph Feichtenhofer · Andrea Vedaldi · João Henriques -
2021 Poster: Unsupervised Part Discovery from Contrastive Reconstruction »
Subhabrata Choudhury · Iro Laina · Christian Rupprecht · Andrea Vedaldi -
2020 Poster: Continuous Surface Embeddings »
Natalia Neverova · David Novotny · Marc Szafraniec · Vasil Khalidov · Patrick Labatut · Andrea Vedaldi -
2020 Poster: Labelling unlabelled videos from scratch with multi-modal self-supervision »
Yuki Asano · Mandela Patrick · Christian Rupprecht · Andrea Vedaldi -
2020 Poster: RELATE: Physically Plausible Multi-Object Scene Synthesis Using Structured Latent Spaces »
Sebastien Ehrhardt · Oliver Groth · Aron Monszpart · Martin Engelcke · Ingmar Posner · Niloy Mitra · Andrea Vedaldi -
2020 Poster: Canonical 3D Deformer Maps: Unifying parametric and non-parametric methods for dense weakly-supervised category reconstruction »
David Novotny · Roman Shapovalov · Andrea Vedaldi -
2019 Poster: PerspectiveNet: A Scene-consistent Image Generator for New View Synthesis in Real Indoor Environments »
David Novotny · Ben Graham · Jeremy Reizenstein -
2019 Poster: Correlated Uncertainty for Learning Dense Correspondences from Noisy Labels »
Natalia Neverova · David Novotny · Andrea Vedaldi