Salvato in:
Dettagli Bibliografici
Autori principali: Sarkar, Meenakshi, Bhardwaj, Devansh, Ghose, Debasish
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2406.14436
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909228199837696
author Sarkar, Meenakshi
Bhardwaj, Devansh
Ghose, Debasish
author_facet Sarkar, Meenakshi
Bhardwaj, Devansh
Ghose, Debasish
contents Stochastic video generation is particularly challenging when the camera is mounted on a moving platform, as camera motion interacts with observed image pixels, creating complex spatio-temporal dynamics and making the problem partially observable. Existing methods typically address this by focusing on raw pixel-level image reconstruction without explicitly modelling camera motion dynamics. We propose a solution by considering camera motion or action as part of the observed image state, modelling both image and action within a multi-modal learning framework. We introduce three models: Video Generation with Learning Action Prior (VG-LeAP) treats the image-action pair as an augmented state generated from a single latent stochastic process and uses variational inference to learn the image-action latent prior; Causal-LeAP, which establishes a causal relationship between action and the observed image frame at time $t$, learning an action prior conditioned on the observed image states; and RAFI, which integrates the augmented image-action state concept into flow matching with diffusion generative processes, demonstrating that this action-conditioned image generation concept can be extended to other diffusion-based models. We emphasize the importance of multi-modal training in partially observable video generation problems through detailed empirical studies on our new video action dataset, RoAM.
format Preprint
id arxiv_https___arxiv_org_abs_2406_14436
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Video Generation with Learned Action Prior
Sarkar, Meenakshi
Bhardwaj, Devansh
Ghose, Debasish
Computer Vision and Pattern Recognition
Robotics
Stochastic video generation is particularly challenging when the camera is mounted on a moving platform, as camera motion interacts with observed image pixels, creating complex spatio-temporal dynamics and making the problem partially observable. Existing methods typically address this by focusing on raw pixel-level image reconstruction without explicitly modelling camera motion dynamics. We propose a solution by considering camera motion or action as part of the observed image state, modelling both image and action within a multi-modal learning framework. We introduce three models: Video Generation with Learning Action Prior (VG-LeAP) treats the image-action pair as an augmented state generated from a single latent stochastic process and uses variational inference to learn the image-action latent prior; Causal-LeAP, which establishes a causal relationship between action and the observed image frame at time $t$, learning an action prior conditioned on the observed image states; and RAFI, which integrates the augmented image-action state concept into flow matching with diffusion generative processes, demonstrating that this action-conditioned image generation concept can be extended to other diffusion-based models. We emphasize the importance of multi-modal training in partially observable video generation problems through detailed empirical studies on our new video action dataset, RoAM.
title Video Generation with Learned Action Prior
topic Computer Vision and Pattern Recognition
Robotics
url https://arxiv.org/abs/2406.14436