AttentionMask: Attentive, Efficient Object Proposal Generation Focusing on Small Objects

Wilms, Christian; Frintrop, Simone

doi:10.1007/978-3-030-20890-5_43

Cited by 19 publications

(68 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It can lay the foundation for consequent target detection, and improve the accuracy and efficiency of the detection. Experimental on the LEVIR dataset [25] demonstrates that the proposed approach holds a 2%∼3% higher recall rate, compared with those of the state-of-the-art methods, including RPN [1], FastMask [12], AttentionMask [24], etc.…”

Section: Introductionmentioning

confidence: 96%

See 1 more Smart Citation

A Multi-Scale Spatial Attention Region Proposal Network for High-Resolution Optical Remote Sensing Imagery

et al. 2021

View full text Add to dashboard Cite

Deep convolutional neural networks (DCNNs) are driving progress in object detection of high-resolution remote sensing images. Region proposal generation, as one of the key steps in object detection, has also become the focus of research. High-resolution remote sensing images usually contain various sizes of objects and complex background, small objects are easy to miss or be mis-identified in object detection. If the recall rate of region proposal of small objects and multi-scale objects can be improved, it will bring an improvement on the performance of the accuracy in object detection. Spatial attention is the ability to focus on local features in images and can improve the learning efficiency of DCNNs. This study proposes a multi-scale spatial attention region proposal network (MSA-RPN) for high-resolution optical remote sensing imagery. The MSA-RPN is an end-to-end deep learning network with a backbone network of ResNet. It deploys three novel modules to fulfill its task. First, the Scale-specific Feature Gate (SFG) focuses on features of objects by processing multi-scale features extracted from the backbone network. Second, the spatial attention-guided model (SAGM) obtains spatial information of objects from the multi-scale attention maps. Third, the Selective Strong Attention Maps Model (SSAMM) adaptively selects sliding windows according to the loss values from the system’s feedback, and sends the windowed samples to the spatial attention decoder. Finally, the candidate regions and their corresponding confidences can be obtained. We evaluate the proposed network in a public dataset LEVIR and compare with several state-of-the-art methods. The proposed MSA-RPN yields a higher recall rate of region proposal generation, especially for small targets in remote sensing images.

show abstract

Section: Introductionmentioning

confidence: 96%

“…For natural images, some approaches adopted attention mechanism to generate proposals specifically for small targets. Christuan Wihms et al [24] designed AttentionMask based on FastMask [12]. They added an extra module to the backbone network of Fast-Mask to focus on the small objects that might be missed before.…”

Section: Introductionmentioning

confidence: 99%

A Multi-Scale Spatial Attention Region Proposal Network for High-Resolution Optical Remote Sensing Imagery

et al. 2021

View full text Add to dashboard Cite

show abstract

“…This leads to better generalization to unseen classes [27]. Since the emergence of deep learning, a number of CNN-based methods for object proposal generation were proposed [29,30,16,37,8,21]. However, most of these systems suffer from imprecise segmentations, especially systems that propose segmentation masks as visible in Fig.…”

Section: Introductionmentioning

confidence: 99%

“…A major reason for the imprecise segmentations is the CNN, which leads to a low resolution in the system's segmentation stage. For instance, in [37] segmentations are based on 10 × 10 feature maps, independent of the object's size. However, those low resolution feature maps are semantically rich and important for the systems' overall success.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

DeepFH Segmentations for Superpixel-based Object Proposal Refinement

Wilms¹,

Frintrop²

2021

Preprint

Self Cite

View full text Add to dashboard Cite

Class-agnostic object proposal generation is an important first step in many object detection pipelines. However, object proposals of modern systems are rather inaccurate in terms of segmentation and only roughly adhere to object boundaries. Since typical refinement steps are usually not applicable to thousands of proposals, we propose a superpixel-based refinement system for object proposal generation systems. Utilizing precise superpixels and superpixel pooling on deep features, we refine initial coarse proposals in an end-to-end learned system. Furthermore, we propose a novel DeepFH segmentation, which enriches the classic Felzenszwalb and Huttenlocher (FH) segmentation with deep features leading to improved segmentation results and better object proposal refinements. On the COCO dataset with LVIS annotations, we show that our refinement based on DeepFH superpixels outperforms state-of-the-art methods and leads to more precise object proposals.

show abstract

Small, but Important: Traffic Light Proposals for Detecting Small Traffic Lights and Beyond

Sanitz,

Wilms,

Frintrop

2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

AttentionMask: Attentive, Efficient Object Proposal Generation Focusing on Small Objects

Cited by 19 publications

References 43 publications

A Multi-Scale Spatial Attention Region Proposal Network for High-Resolution Optical Remote Sensing Imagery

A Multi-Scale Spatial Attention Region Proposal Network for High-Resolution Optical Remote Sensing Imagery

DeepFH Segmentations for Superpixel-based Object Proposal Refinement

Small, but Important: Traffic Light Proposals for Detecting Small Traffic Lights and Beyond

Contact Info

Product

Resources

About