Dual Adversarial Network for Unsupervised Ground/Satellite-to-Aerial Scene Adaptation

Lin, Jianzhe; Mou, Lichao; Yu, Tianze; Zhu, Xiao Xiang; Wang, Z. Jane

doi:10.1145/3394171.3413893

Cited by 10 publications

(7 citation statements)

References 57 publications

(40 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The satellite/aerial to ground matching at the scene level is highly challenging due to the large semantic gap between the ground and overhead scene, but still resolvable. For example, the authors in [39] proposed an dual adversarial solution for unsupervised satellite/aerial to ground scene adaptation solution. However, it becomes very crucial when object level matching is concerned, e.g., when approaching automatic geolocalization [40], [41], and in particular when considering image synthesis [42]- [44], where strong geometric models of the various modalities, with the ability to model uncertainty, are strongly needed.…”

Section: Multimodal Approaches With Social Mediamentioning

confidence: 99%

Toward a Collective Agenda on AI for Earth Science Data Analysis

Tuia¹,

Roscher

Wegner

et al. 2021

IEEE Geosci. Remote Sens. Mag.

Self Cite

View full text Add to dashboard Cite

This is the pre-acceptance version, to read the final version published in the Geoscience and Remote Sensing Magazine, please go to: 10.1109/MGRS.2020.3043504 In the last years we have witnessed the fields of geosciences and remote sensing and artificial intelligence to become closer. Thanks to both the massive availability of observational data, improved simulations, and algorithmic advances, these disciplines have found common objectives and challenges to advance the modeling and understanding of the Earth system. Despite such great opportunities, we also observed a worrying tendency to remain in disciplinary comfort zones applying recent advances from artificial intelligence on well resolved remote sensing problems. Here we take a position on research directions where we think the interface between these fields will have the most impact and become potential game changers. In our declared agenda for AI on Earth sciences, we aim to inspire researchers, especially the younger generations, to tackle these challenges for a real advance of remote sensing and the geosciences.

show abstract

Section: Multimodal Approaches With Social Mediamentioning

confidence: 99%

Toward a Collective Agenda on AI for Earth Science Data Analysis

Tuia¹,

Roscher

Wegner

et al. 2021

IEEE Geosci. Remote Sens. Mag.

Self Cite

View full text Add to dashboard Cite

show abstract

“…In recent years, many efforts ( Zhu et al, 2017 ), e.g., developing novel network architectures ( Murray et al, 2019 , Cheng et al, 2020 , Bi et al, 2020 , Niazmardi et al, 2017 , Lin et al, 2020 , Zhu et al, 2018 ) and pipelines ( Byju et al, 2000 , Xu et al, 2020 , Wang et al, 2019 , Zhu et al, 2019 ), publishing large-scale datasets ( Xia et al, 2017 , Jin et al, 2018 ), introducing multi-modal and multi-temporal data ( Hu et al, 2020 , Tuia et al, 2016 , Ru et al, 2020 , Li et al, 2020a ), have been deployed to address this task, and most of them treat it as a single-label classification problem. A common assumption shared by these researches is that an aerial image belongs to only one scene category, while in real-world scenarios, it is more often that there exist various scenes in a single image (cf.…”

Section: Introductionmentioning

confidence: 99%

Aerial scene understanding in the wild: Multi-scene recognition via prototype-based memory networks

Hua

Mou

Lin

et al. 2021

ISPRS Journal of Photogrammetry and Remote Sensing

Self Cite

View full text Add to dashboard Cite

Aerial scene recognition is a fundamental visual task and has attracted an increasing research interest in the last few years. Most of current researches mainly deploy efforts to categorize an aerial image into one scene-level label, while in real-world scenarios, there often exist multiple scenes in a single image. Therefore, in this paper, we propose to take a step forward to a more practical and challenging task, namely multi-scene recognition in single images. Moreover, we note that manually yielding annotations for such a task is extraordinarily time- and labor-consuming. To address this, we propose a prototype-based memory network to recognize multiple scenes in a single image by leveraging massive well-annotated single-scene images. The proposed network consists of three key components: 1) a prototype learning module, 2) a prototype-inhabiting external memory, and 3) a multi-head attention-based memory retrieval module. To be more specific, we first learn the prototype representation of each aerial scene from single-scene aerial image datasets and store it in an external memory. Afterwards, a multi-head attention-based memory retrieval module is devised to retrieve scene prototypes relevant to query multi-scene images for final predictions. Notably, only a limited number of annotated multi-scene images are needed in the training phase. To facilitate the progress of aerial scene recognition, we produce a new multi-scene aerial image (MAI) dataset. Experimental results on variant dataset configurations demonstrate the effectiveness of our network. Our dataset and codes are publicly available 1 .

show abstract

“…Besides, it assigns static weights to neighbours, which results in the inability to adapt to different inference relations. We argue that the accuracy of the inference results is not only affected by the explicit characteristics of the entity itself, but also by the implicit characteristics [9, 10]. For example, the implicit characteristics of the entity's neighbours can cascade to affect the overall characteristics of the entities [11, 12].…”

Section: Introductionmentioning

confidence: 99%

Hybrid attention mechanism for few‐shot relational learning of knowledge graphs

Guo

et al. 2021

IET Computer Vision

View full text Add to dashboard Cite

Few-shot knowledge graph (KG) reasoning is the main focus in the field of knowledge graph reasoning. In order to expand the application fields of the knowledge graph, a large number of studies are based on a large number of training samples. However, we have learnt that there are actually many missing relationships or entities in the knowledge graph, and in most cases, there are not many training instances when implementing new relationships. To tackle it, in this study, the authors aim to predict a new entity given few reference instances, even only one training instance. A few-shot learning framework based on a hybrid attention mechanism is proposed. The framework employs traditional embedding models to extract knowledge, and uses an attenuated attention network and a self-attention mechanism to obtain the hidden attributes of entities. Thus, it can learn a matching metric by considering both the learnt embeddings and one-hop graph structures. The experimental results present that the model has achieved significant performance improvements on the NELL-One and Wiki-One datasets.

show abstract

Dual Adversarial Network for Unsupervised Ground/Satellite-to-Aerial Scene Adaptation

Cited by 10 publications

References 57 publications

Toward a Collective Agenda on AI for Earth Science Data Analysis

Toward a Collective Agenda on AI for Earth Science Data Analysis

Aerial scene understanding in the wild: Multi-scene recognition via prototype-based memory networks

Hybrid attention mechanism for few‐shot relational learning of knowledge graphs

Contact Info

Product

Resources

About