FMSSD: Feature-Merged Single-Shot Detection for Multiscale Objects in Large-Scale Remote Sensing Imagery

Wang, Peijin; Sun, Xian; Diao, Wenhui; Fu, Kun

doi:10.1109/tgrs.2019.2954328

Cited by 239 publications

(90 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The effectiveness of context information has been verified by many studies [43][44][45] in aerial object detection, especially for small objects or occluded objects. The intuition behind these works is that low-level high-resolution features from shallow layers favor localization, while high-level low-resolution features from deeper layers help classification.…”

Section: Context Information In Rs Object Detectionmentioning

confidence: 85%

“…Accordingly, feature pyramid network (FPN) [38] and deconvolutional single shot detector (DSSD) [39] take advantage of lateral connections to combine low-resolution semantically strong features with high-resolution semantically weak features, yielding enriched feature maps at all scales. Driven by the power of FPN in multi-scale detection, an atrous spatial feature pyramid (ASFP) [44] used atrous convolution layers at different rates for more effective fusion of multi-scale context information. Alternatively, image cascade network (ICN) [46] combined image cascade and FPN to allow extracting features at different levels and scales.…”

Section: Context Information In Rs Object Detectionmentioning

confidence: 99%

“…We compared our proposed F 3 -Net with 10 state-of-the-art approaches, including one-stage methods, such as SSD [13], YOLOV2 [54], and FMSSD [44], and two-stage methods, for example, SCRDet [41], FR-O [2], ICN [46], RoI-Transformer [20], and APE [24]. Tables 2 and 3 report the quantitative comparison between the proposed F 3 -Net and several state-of-the-art approaches on both OBB and HBB tasks, respectively.…”

Section: Dota Datasetmentioning

confidence: 99%

See 2 more Smart Citations

ℱ3-Net: Feature Fusion and Filtration Network for Object Detection in Optical Remote Sensing Images

Xiong

et al. 2020

Remote Sensing

View full text Add to dashboard Cite

Object detection in remote sensing (RS) images is a challenging task due to the difficulties of small size, varied appearance, and complex background. Although a lot of methods have been developed to address this problem, many of them cannot fully exploit multilevel context information or handle cluttered background in RS images either. To this end, in this paper, we propose a feature fusion and filtration network (F3-Net) to improve object detection in RS images, which has higher capacity of combining the context information at multiple scales while suppressing the interference from the background. Specifically, F3-Net leverages a feature adaptation block with a residual structure to adjust the backbone network in an end-to-end manner, better considering the characteristics of RS images. Afterward, the network learns the context information of the object at multiple scales by hierarchically fusing the feature maps from different layers. In order to suppress the interference from cluttered background, the fused feature is then projected into a low-dimensional subspace by an additional feature filtration module. As a result, more relevant and accurate context information is extracted for further detection. Extensive experiments on DOTA, NWPU VHR-10, and UCAS AOD datasets demonstrate that the proposed detector achieves very promising detection performance.

show abstract

Section: Context Information In Rs Object Detectionmentioning

confidence: 85%

Section: Context Information In Rs Object Detectionmentioning

confidence: 99%

Section: Dota Datasetmentioning

confidence: 99%

See 1 more Smart Citation

ℱ3-Net: Feature Fusion and Filtration Network for Object Detection in Optical Remote Sensing Images

Xiong

et al. 2020

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…S PACEBORNE and airborne remotely sensed images offer an important tool to deal with current societal needs as well as future challenges [1]. From the study of the spectral properties of the Earth surface [2]- [4], through the visual detection of specific targets [5]- [7], to planning and monitoring of land-cover [6], [8]- [10], there are multiple domains where remote sensing (RS) images become particularly useful, and the growing development of different Earth observation (EO) missions exemplifies this fact [11]. As a result, recent years have witnessed an explosive growth in RS image collections, aimed at implementing big-scale operational services which demand new efficient methodologies to manage and retrieve relevant information from the massive resulting RS archives [10], [12]- [14].…”

Section: Introductionmentioning

confidence: 99%

Deep Hashing Based on Class-Discriminated Neighborhood Embedding

Kang

Fernández-Beltrán

et al. 2020

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Deep-hashing methods have drawn significant attention during the past years in the field of Remote Sensing (RS) owing to their prominent capabilities for capturing the semantics from complex RS scenes and generating the associated hash codes in an end-to-end manner. Most existing deep-hashing methods exploit pairwise and triplet losses to learn the hash codes with the preservation of semantic-similarities which require the construction of image pairs and triplets based on supervised information (e.g. class labels). However, the learned Hamming spaces based on these losses may not be optimal due to an insufficient sampling of image pairs and triplets for scalable RS archives. To solve this limitation, we propose a new deephashing technique based on the class-discriminated neighborhood embedding (CDNE), which can properly capture the locality structures among the RS scenes and distinguish images classwisely in the Hamming space. An extensive experimentation has been conducted in order to validate the effectiveness of the proposed method by comparing it with several state-of-the-art conventional and deep-hashing methods. The related codes of this article will be made publicly available for reproducible research by the community.

show abstract

“…It has a significant impact on Land Use and Land Cover (LULC) determination (Zhang et al, 2013, Zhu et al, 2016, vegetation mapping (Li, Shao, 2013, Mishra, Crews, 2014, urban planning and so on. While it also offers a foundation for semantic segmentation (Kampffmeyer et al, 2016), object detection (Wang et al, 2019a, Fu et al, 2020, Feng et al, 2019, Fine Grained Visual Classification (FGVC) (Fu et al, 2019) and other extension tasks. Due to the rapidly increasing quantity of remote sensing images, highly complex geometric structures and quite large scale images (Zhao et al, 2016), how to improve a model's performance and increase its automation of design procedure are still tricky problems.…”

Section: Introductionmentioning

confidence: 99%

Automated and Lightweight Network Design via Random Search for Remote Sensing Image Scene Classification

Diao

Sun

et al. 2020

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

Abstract. With the development of deep learning, remote sensing image scene classification technology has been greatly improved. However, current deep networks used for scene classification usually introduce ingenious extra modules to fit the characteristics of remote sensing images. It causes a high labor cost and brings more parameters, which makes the network more complicated and poses new intractable problems. In this paper, we rethink this popular “add module” pattern and propose a more lightweight model, called ProbDenseNet (PDN). PDN is obtained via a random search strategy in Neural Architecture Search (NAS) which is an automated network design manner. In our method, all topological connections are assigned importance degrees which subject to a uniform distribution. And we set a regulator to adjust the sparsity of the network. By this way, the design procedure is more automated and the network structure becomes more lightweight. Experimental results on AID benchmark demonstrate that the proposed PDN model can achieve competitive performance even with much fewer parameters. And we also find that excessive connections do not always improve the network’s performance while they can drag down the network’s behavior as well. Furthermore, we conduct experiments on Vaihingen dataset with classical Fully Convolutional Network (FCN) framework. Quantitative and qualitative results both indicate that the features learned by PDN can also transfer in semantic segmentation task.

show abstract

FMSSD: Feature-Merged Single-Shot Detection for Multiscale Objects in Large-Scale Remote Sensing Imagery

Cited by 239 publications

References 43 publications

ℱ3-Net: Feature Fusion and Filtration Network for Object Detection in Optical Remote Sensing Images

ℱ3-Net: Feature Fusion and Filtration Network for Object Detection in Optical Remote Sensing Images

Deep Hashing Based on Class-Discriminated Neighborhood Embedding

Automated and Lightweight Network Design via Random Search for Remote Sensing Image Scene Classification

Contact Info

Product

Resources

About