Bridging Global Context Interactions for High-Fidelity Pluralistic Image Completion
IEEE Transactions on Pattern Analysis and Machine Intelligence · 2024-12 · 46 (12) · 8320-8333
DOI: 10.1109/tpami.2024.3403695 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024-12. Verfasst von Chuanxia Zheng und 5 weiteren Autor:innen. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 81 Literaturangaben.
Abstract (Originalsprache)
We introduce PICFormer, a novel framework for Pluralistic Image Completion using a transFormer based architecture, that achieves both high quality and diversity at a much faster inference speed. Our key contribution is to introduce a code-shared codebook learning using a restrictive CNN on small and non-overlapping receptive fields (RFs) for the local visible token representation. This results in a compact yet expressive discrete representation, facilitating efficient modeling of global visible context relations by the transformer. Unlike the prevailing autoregressive approaches, we proposed to sample all tokens simultaneously, leading to more than 100× faster inference speed. To enhance appearance consistency between visible and generated regions, we further propose a novel attention-aware layer (AAL), designed to better exploit distantly related high-frequency features. Through extensive experiments, we demonstrate that the PICFormer efficiently learns semantically-rich discrete codes, resulting in significantly improved image quality. Moreover, our diverse image completion framework surpasses State-of-the-Art methods on multiple image completion datasets.
Zitieren
APA
Zheng, C., Song, G., Cham, T. J., Cai, J., Luo, L., & Phung, D. (2024). Bridging Global Context Interactions for High-Fidelity Pluralistic Image Completion. IEEE Transactions on Pattern Analysis and Machine Intelligence, 46(12), 8320-8333. https://doi.org/10.1109/tpami.2024.3403695
Vancouver
Zheng C, Song G, Cham TJ, Cai J, Luo L, Phung D. Bridging Global Context Interactions for High-Fidelity Pluralistic Image Completion. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2024;46(12):8320-8333. doi:10.1109/tpami.2024.3403695
BibTeX
@article{Zheng2024,
author = {Zheng, Chuanxia and Song, Guoxian and Cham, Tat-Jen and Cai, Jianfei and Luo, Linjie and Phung, Dinh},
title = {Bridging Global Context Interactions for High-Fidelity Pluralistic Image Completion},
journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
year = {2024},
volume = {46},
number = {12},
pages = {8320--8333},
doi = {10.1109/tpami.2024.3403695}
}
EUVIMED