PistonNet: Object Separating From Background by Attention for Weakly Supervised Ship Detection

Yi, Yang; Pan, Zongxu; Hu, Yuxin; Ding, Chibiao

doi:10.1109/jstars.2022.3184637

Cited by 7 publications

(4 citation statements)

References 30 publications

(14 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, another interesting observation concerns PistonNet [21]. In particular, it can be observed that the performance of this method on NWPU VHR-10 is 11.38% lower than MPFP-Net [43].…”

Section: Other Performancesmentioning

confidence: 93%

“…In recent years, the RS domain has been gaining more interest and novel methods have been continuously proposed. In particular, WSOD offers many possibilities for applications that otherwise would not be possible due to the large number of manual annotations required, e.g., vehicle detection [19][20][21], marine animal detection [22], or defective insulator detection [23]. We introduce a timeline to precisely define the evolution of RSWSOD methods and describe the pros and cons and the challenges of each state-of-theart method.…”

Section: Focus Of the Surveymentioning

confidence: 99%

“…More recently, Yang et al [21] addressed the task of ship detection, exploiting an image transformer [82] called PistonNet. PistonNet proposes the introduction of an artificial point in the feature map, whose aim is to bring the background values to 0 while keeping the important area to 1 to allow a clear separation between the two.…”

Section: Other Dl-basedmentioning

confidence: 99%

“…Transformer-based architectures: Transformers were born to tackle natural language processing (NLP) problems, but their usage has gained much attention in the CV field due to their powerful capabilities. PistonNet [21] showed interesting results with the use of image transformers [82] and provided good generalization capabilities, despite being developed for the specific use-case of ship detection. This powerful family of architectures could be extended to general-purpose RSWSOD.…”

mentioning

confidence: 99%

See 3 more Smart Citations

Weakly Supervised Object Detection for Remote Sensing Images: A Survey

et al. 2022

View full text Add to dashboard Cite

The rapid development of remote sensing technologies and the availability of many satellite and aerial sensors have boosted the collection of large volumes of high-resolution images, promoting progress in a wide range of applications. As a consequence, Object detection (OD) in aerial images has gained much interest in the last few years. However, the development of object detectors requires a massive amount of carefully labeled data. Since annotating datasets is very time-consuming and may require expert knowledge, a consistent number of weakly supervised object localization (WSOL) and detection (WSOD) methods have been developed. These approaches exploit only coarse-grained metadata, typically whole image labels, to train object detectors. However, many challenges remain open due to the missing location information in the training process of WSOD approaches and to the complexity of remote sensing images. Furthermore, methods studied for natural images may not be directly applicable to remote sensing images (RSI) and may require carefully designed adaptations. This work provides a comprehensive survey of the recent achievements of remote sensing weakly supervised object detection (RSWSOD). An analysis of the challenges related to RSWSOD is presented, the advanced techniques developed to improve WSOD are summarized, the available benchmarking datasets are described and a discussion of future directions of RSWSOD research is provided.

show abstract

“…Finally, another interesting observation concerns PistonNet [21]. In particular, it can be observed that the performance of this method on NWPU VHR-10 is 11.38% lower than MPFP-Net [43].…”

Section: Other Performancesmentioning

confidence: 93%