Saved in:
Bibliographic Details
Main Authors: Traub, Manuel, Becker, Frederic, Otte, Sebastian, Butz, Martin V.
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2310.10372
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909166291910656
author Traub, Manuel
Becker, Frederic
Otte, Sebastian
Butz, Martin V.
author_facet Traub, Manuel
Becker, Frederic
Otte, Sebastian
Butz, Martin V.
contents While human infants exhibit knowledge about object permanence from two months of age onwards, deep-learning approaches still largely fail to recognize objects' continued existence. We introduce a slot-based autoregressive deep learning system, the looped location and identity tracking model Loci-Looped, which learns to adaptively fuse latent imaginations with pixel-space observations into consistent latent object-specific what and where encodings over time. The novel loop empowers Loci-Looped to learn the physical concepts of object permanence, directional inertia, and object solidity through observation alone. As a result, Loci-Looped tracks objects through occlusions, anticipates their reappearance, and shows signs of surprise and internal revisions when observing implausible object behavior. Notably, Loci-Looped outperforms state-of-the-art baseline models in handling object occlusions and temporary sensory interruptions while exhibiting more compositional, interpretable internal activity patterns. Our work thus introduces the first self-supervised interpretable learning model that learns about object permanence directly from video data without supervision.
format Preprint
id arxiv_https___arxiv_org_abs_2310_10372
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Learning Object Permanence from Videos via Latent Imaginations
Traub, Manuel
Becker, Frederic
Otte, Sebastian
Butz, Martin V.
Computer Vision and Pattern Recognition
While human infants exhibit knowledge about object permanence from two months of age onwards, deep-learning approaches still largely fail to recognize objects' continued existence. We introduce a slot-based autoregressive deep learning system, the looped location and identity tracking model Loci-Looped, which learns to adaptively fuse latent imaginations with pixel-space observations into consistent latent object-specific what and where encodings over time. The novel loop empowers Loci-Looped to learn the physical concepts of object permanence, directional inertia, and object solidity through observation alone. As a result, Loci-Looped tracks objects through occlusions, anticipates their reappearance, and shows signs of surprise and internal revisions when observing implausible object behavior. Notably, Loci-Looped outperforms state-of-the-art baseline models in handling object occlusions and temporary sensory interruptions while exhibiting more compositional, interpretable internal activity patterns. Our work thus introduces the first self-supervised interpretable learning model that learns about object permanence directly from video data without supervision.
title Learning Object Permanence from Videos via Latent Imaginations
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2310.10372