Real-Time Pedestrian Detection Using Enhanced Representations from Light-Weight YOLO Network

Bagi, Shayan Shirahmad Gale; Moshiri, Behzad; Garakani, Hossein Gharaee; Crowley, Mark; Mehrannia, Pouya

doi:10.1109/codit55151.2022.9804060

Cited by 3 publications

(4 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Today, real-time methods for target recognition such as You Only Look Once (YOLO), a network called the "single pass network", are preferred in conditions where the latency of the algorithm is relevant. YOLO reduces processing time compared to other ML techniques: the network was created for fast object recognition in images or videos, like in [34], and was later applied to Radar signals. In [35], a Multiple Input Multiple Output (MIMO) Radar able to exploit a bidimensional array is used to obtain an image similar to one obtained from an RGB sensor.…”

Section: A State-of-the-artmentioning

confidence: 99%

See 1 more Smart Citation

mmDetect: YOLO-based Processing of mm-Wave Radar Data for Detecting Moving People

Raimondi,

Ciattaglia,

Nocera

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

Section: A State-of-the-artmentioning

confidence: 99%

“…For the tests, we consider a YOLOv3 [34] and a SqueezeNet as backbone CNN. SqueezeNet is a pre-trained model on Im-ageNet [51], to which layer freezing and transfer learning are applied, and represents one of the most lightweight solutions in terms of complexity and the number of parameters.…”

Section: B Yolo Parametersmentioning

confidence: 99%

mmDetect: YOLO-based Processing of mm-Wave Radar Data for Detecting Moving People

Raimondi,

Ciattaglia,

Nocera

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…YOLO reduces processing time compared to other ML techniques. The network was created for object recognition on images or videos, like in [33], and was later applied to Radar signals. In [34], a MIMO Radar able to exploit a bidimensional array is used to obtain an image similar to one obtained from an RGB sensor.…”

Section: A State-of-the-artmentioning

confidence: 99%

mmDetect: YOLO-Based Processing of mm-Wave Radar Data for Detecting Moving People

Raimondi,

Ciattaglia,

Nocera

et al. 2024

IEEE Sensors J.

View full text Add to dashboard Cite

The terms and conditions for the reuse of this version of the manuscript are specified in the publishing policy. The use of copyrighted works requires the consent of the rights' holder (author or publisher). Works made available under a Creative Commons license or a Publisher's custom-made license can be used according to the terms and conditions contained therein. See editor's website for further information and terms and conditions. This item was downloaded from IRIS Università Politecnica delle Marche (https://iris.univpm.it). When citing, please refer to the published version.

show abstract

“…Results on the whole 84 Vysis images were 83.91%, and 87.41% on the set with 18 low-quality images removed. Redmon et al proposed a one-stage architecture to obtain the category probability and location coordinates, eliminating the need for region proposals [22] and used in breast cancer detection [23], nucleus detection for tumor [24], and pedestrian detection [25] [26]. Bai et al devised a three-step architecture in 2020 by employing UNET and YOLOv3 network [27].…”

Section: Introductionmentioning

confidence: 99%

Fully Automatic Karyotyping via Deep Convolutional Neural Networks

Wang,

Yu,

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Objective: Chromosome karyotyping is an important yet labor-intensive procedure for diagnosing genetic diseases. Automating such a procedure drastically reduces the manual work of cytologists and increases congenital disease diagnosing precision. Researchers have contributed to chromosome segmentation and classification for decades. However, very few studies integrate the two tasks as a unified, fully automatic procedure or achieved a promising performance. Methods: This paper addresses the gap by presenting: (1) A novel chromosome segmentation module named ChrRender, with the idea of rendering the chromosome instances by combining rich global features from the backbone and coarse mask prediction from Mask R-CNN. (2) A devised chromosome classification module named ChrNet4 that pays more attention to channel-wise dependencies from aggregated informative features and calibrating the channel interdependence. (3) An integrated Render-Attention-Architecture to accomplish fully automatic karyotyping with segmentation and classification modules. (4) A strategy for eliminating differences between training data and segmentation output data to be classified. Results: These proposed methods are implemented in three ways on the public Q-band BioImLab dataset and a G-band private dataset. The results indicate promising performance: (1) on the joint karyotyping task, which predicts a karyotype image by first segmenting an original microscopical image, then classifying each segmentation output with a precision of 89.75% and 94.22% on the BioImLab and private dataset, respectively; (2) On the separate task with two datasets, ChrRender obtained AP50 of 96.652% and 96.809% for segmentation, ChrNet4 achieved 95.24% and 94.07% for classification, respectively. The COCO format annotation files of BioImLab used in this paper are available at https://github.com/Alex17swim/BioImLab Conclusions: The study introduces an integrated workflow to predict a karyotyping image from a Microscopical Chromosome Image. With state-of-the-art performance on a public dataset, the proposed Render-Attention-Architecture has accomplished fully automatic chromosome karyotyping.

show abstract

Real-Time Pedestrian Detection Using Enhanced Representations from Light-Weight YOLO Network

Cited by 3 publications

References 19 publications

mmDetect: YOLO-based Processing of mm-Wave Radar Data for Detecting Moving People

mmDetect: YOLO-based Processing of mm-Wave Radar Data for Detecting Moving People

mmDetect: YOLO-Based Processing of mm-Wave Radar Data for Detecting Moving People

Fully Automatic Karyotyping via Deep Convolutional Neural Networks

Contact Info

Product

Resources

About