Object Recognition and Tracking Using Bayesian Networks for Augmented Reality Systems

Silva, Rute; Rodrigues, Paulo S.; Giraldi, Gilson A.; Cunha, Gerson Gomes

doi:10.1109/iv.2005.92

Cited by 2 publications

(3 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Findings confirmed that our proposed model was stable for vehicular detection on the aerial and K-road datasets. In addition, we compared the detection results of recent CNN-based object detectors, namely Faster R-CNN [46] with the use of the Z and F model, Faster R-CNN [46] with the use of the VGG-16 model, and Fast R-CNN [51] with the use of the VGG-model for the VEDAI dataset.…”

Section: Detection Results and Performance Analysismentioning

confidence: 99%

“…Prior research on object detection has focused on template-matching and part-based models. Subsequently, research focused on statistical classifiers, such as support vector machines [44], AdaBoost [45], Bayes' theorem [46], decision trees [47], K-nearest neighbors [48], and random forest techniques [49]. All of these are initial object detectors based on statistical classifiers.…”

Section: Deep Learning-based Object-detection Modelsmentioning

confidence: 99%

See 1 more Smart Citation

SRODNet: Object Detection Network Based on Super Resolution for Autonomous Vehicles

2022

View full text Add to dashboard Cite

Object detection methods have been applied in several aerial and traffic surveillance applications. However, object detection accuracy decreases in low-resolution (LR) images owing to feature loss. To address this problem, we propose a single network, SRODNet, that incorporates both super-resolution (SR) and object detection (OD). First, a modified residual block (MRB) is proposed in the SR to recover the feature information of LR images, and this network was jointly optimized with YOLOv5 to benefit from hierarchical features for small object detection. Moreover, the proposed model focuses on minimizing the computational cost of network optimization. We evaluated the proposed model using standard datasets such as VEDAI-VISIBLE, VEDAI-IR, DOTA, and Korean highway traffic (KoHT), both quantitatively and qualitatively. The experimental results show that the proposed method improves the accuracy of vehicular detection better than other conventional methods.

show abstract

Section: Detection Results and Performance Analysismentioning

confidence: 99%

Section: Deep Learning-based Object-detection Modelsmentioning

confidence: 99%

SRODNet: Object Detection Network Based on Super Resolution for Autonomous Vehicles

2022

View full text Add to dashboard Cite

show abstract

“…In essence, SLAM represents a method employed in the fields of robotics and computer vision, enabling a robot to simultaneously construct a map of an unfamiliar space and determine its location within that space [11]. Conversely, AR is a technology that overlays digital data onto the physical world [12]. By integrating SLAM and AR, a range of fascinating applications can emerge.…”

Section: Some Examples Of the Combination Of Slam And Ar Technologymentioning

confidence: 99%

Application of SLAM technology combined with AR in minimally invasive surgery

Yang,

Zhou

2023

ACE

View full text Add to dashboard Cite

Minimally invasive surgery (MIS) is revolutionizing the medical field, offering patients faster recovery times and reduced risk of complications. However, MIS demands precise navigation and visualization of the surgical site, which can be challenging due to the limited field of view and constrained working space. In this study, we investigate the integration of simultaneous localization and mapping (SLAM) technology with augmented reality (AR) to enhance the capabilities of MIS and improve surgical outcomes. SLAM technology enables real-time 3D mapping and tracking of the surgical environment, while AR overlays crucial information, such as anatomical structures and surgical guidance, onto the surgeon's view. The proposed system combines these technologies to create a comprehensive, interactive representation of the patient's anatomy, providing surgeons with a more intuitive and accurate understanding of the surgical site. We assess the performance of the integrated SLAM-AR system based on a review of relevant literature, focusing on its accuracy, usability, and impact on surgical efficiency. The findings suggest that the SLAM-AR system has the potential to significantly improve spatial awareness and navigation during MIS, potentially reducing procedure times and complication rates. Furthermore, the system shows promise for enhancing surgeon training and promoting the adoption of MIS techniques in various surgical fields.

show abstract

Object Recognition and Tracking Using Bayesian Networks for Augmented Reality Systems

Cited by 2 publications

References 19 publications

SRODNet: Object Detection Network Based on Super Resolution for Autonomous Vehicles

SRODNet: Object Detection Network Based on Super Resolution for Autonomous Vehicles

Application of SLAM technology combined with AR in minimally invasive surgery

Contact Info

Product

Resources

About