Many-Shot from Low-Shot: Learning to Annotate Using Mixed Supervision for Object Detection

Biffi, Carlo; McDonagh, Steven; Torr, Philip H. S.; Leonardis, Aleš; Parisot, Sarah

doi:10.1007/978-3-030-58598-3_3

Cited by 9 publications

(3 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The paper also proposed criteria for gathering future datasets. Biffi et al discussed the challenges of object detection relying on time-consuming manual annotations [146] They introduced an online annotation module (OAM) that generates reliable annotations from weakly labeled images. This OAM can enhance the performance of Fast(er) R-CNN, improving mAP by 17% and AP50 by 9% on PASCAL VOC 2007 and MS-COCO benchmarks, outperforming other methods using mixed supervision.…”

Section: A Annotating Training Datamentioning

confidence: 99%

Synchronizing Object Detection: Applications, Advancements and Existing Challenges

Tanzib Hosain,

Zaman,

Abir

et al. 2024

IEEE Access

View full text Add to dashboard Cite

From pivotal roles in autonomous vehicles, healthcare diagnostics, and surveillance systems to seamlessly integrating with augmented reality, object detection algorithms stand as the cornerstone in unraveling the complexities of the visual world. Tracing the trajectory from conventional region-based methods to the latest neural network architectures reveals a technological renaissance where algorithms metamorphose into digital artisans. However, this journey is not without hurdles, prompting researchers to grapple with real-time detection, robustness in varied environments, and interpretability amidst the intricacies of deep learning. The allure of addressing issues such as occlusions, scale variations, and finegrained categorization propels exploration into uncharted territories, beckoning the scholarly community to contribute to an ongoing saga of innovation and discovery. This research offers a comprehensive panorama, encapsulating the applications reshaping our digital reality, the advancements pushing the boundaries of perception, and the open issues extending an invitation to the next generation of visionaries to explore uncharted frontiers within object detection.

show abstract

Section: A Annotating Training Datamentioning

confidence: 99%

Synchronizing Object Detection: Applications, Advancements and Existing Challenges

Tanzib Hosain,

Zaman,

Abir

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…This, however, brings new challenges and solutions in DL from multi-modal data, which leads to flows of efforts to address. In parallel, more and more recent works consider leveraging the vast volume of both nature images in CV and remote sensing images to perform so-called X-shot learning [20][21][22][23][24][25], harnessing the Object detection methods date back to early and traditional methods that rely on hand-crafted and distinctive features, such as SIFT [6] and HOG [7], and match objects of interest in the images based on object examples (template image). Then, bounding boxes are extracted to describe the successful detections.…”

Section: Introductionmentioning

confidence: 99%

“…This, however, brings new challenges and solutions in DL from multi-modal data, which leads to flows of efforts to address. In parallel, more and more recent works consider leveraging the vast volume of both nature images in CV and remote sensing images to perform so-called X-shot learning [20][21][22][23][24][25], harnessing the power of pre-trained and fine-tuned networks to boost object detection performances [25,26]. This was further aided by using the nowadays well-developed language models to realize automated object detection in a much larger label space, i.e., fine-grained object detection.…”

Section: Introductionmentioning

confidence: 99%

Remote Sensing Object Detection in the Deep Learning Era—A Review

Gui,

Song,

Qin

et al. 2024

Remote Sensing

View full text Add to dashboard Cite

Given the large volume of remote sensing images collected daily, automatic object detection and segmentation have been a consistent need in Earth observation (EO). However, objects of interest vary in shape, size, appearance, and reflecting properties. This is not only reflected by the fact that these objects exhibit differences due to their geographical diversity but also by the fact that these objects appear differently in images collected from different sensors (optical and radar) and platforms (satellite, aerial, and unmanned aerial vehicles (UAV)). Although there exists a plethora of object detection methods in the area of remote sensing, given the very fast development of prevalent deep learning methods, there is still a lack of recent updates for object detection methods. In this paper, we aim to provide an update that informs researchers about the recent development of object detection methods and their close sibling in the deep learning era, instance segmentation. The integration of these methods will cover approaches to data at different scales and modalities, such as optical, synthetic aperture radar (SAR) images, and digital surface models (DSM). Specific emphasis will be placed on approaches addressing data and label limitations in this deep learning era. Further, we survey examples of remote sensing applications that benefited from automatic object detection and discuss future trends of the automatic object detection in EO.

show abstract

Active Learning Strategies for Weakly-Supervised Object Detection

Siméoni

Gidaris

et al. 2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Many-Shot from Low-Shot: Learning to Annotate Using Mixed Supervision for Object Detection

Cited by 9 publications

References 25 publications

Synchronizing Object Detection: Applications, Advancements and Existing Challenges

Synchronizing Object Detection: Applications, Advancements and Existing Challenges

Remote Sensing Object Detection in the Deep Learning Era—A Review

Active Learning Strategies for Weakly-Supervised Object Detection

Contact Info

Product

Resources

About