Filters








310 Hits in 4.5 sec

Occlusion-aware R-CNN: Detecting Pedestrians in a Crowd [article]

Shifeng Zhang, Longyin Wen, Xiao Bian, Zhen Lei, Stan Z. Li
2018 arXiv   pre-print
In this paper, we propose a new occlusion-aware R-CNN (OR-CNN) to improve the detection accuracy in the crowd.  ...  Pedestrian detection in crowded scenes is a challenging problem since the pedestrians often gather together and occlude each other.  ...  Conclusions In this paper, we present a new occlusion-aware R-CNN method to improve the pedestrian detection accuracy in crowded scenes.  ... 
arXiv:1807.08407v1 fatcat:rtd2xx7yj5gkjo4hw4y7hyresq

Occlusion-Aware R-CNN: Detecting Pedestrians in a Crowd [chapter]

Shifeng Zhang, Longyin Wen, Xiao Bian, Zhen Lei, Stan Z. Li
2018 Lecture Notes in Computer Science  
In this paper, we propose a new occlusion-aware R-CNN (OR-CNN) to improve the detection accuracy in the crowd.  ...  Pedestrian detection in crowded scenes is a challenging problem since the pedestrians often gather together and occlude each other.  ...  Conclusions In this paper, we present a new occlusion-aware R-CNN method to improve the pedestrian detection accuracy in crowded scenes.  ... 
doi:10.1007/978-3-030-01219-9_39 fatcat:mh3vyhecvfcajkffdh4momkfyq

Adaptive NMS: Refining Pedestrian Detection in a Crowd [article]

Songtao Liu, Di Huang, Yunhong Wang
2019 arXiv   pre-print
Pedestrian detection in a crowd is a very challenging issue.  ...  This paper addresses this problem by a novel Non-Maximum Suppression (NMS) algorithm to better refine the bounding boxes given by detectors.  ...  Therefore, it becomes a necessity to work on pedestrian detection in a crowd.  ... 
arXiv:1904.03629v1 fatcat:y3mfd6zlnbcezmtff7hlg3lkm4

YOLO-FaceV2: A Scale and Occlusion Aware Face Detector [article]

Ziping Yu, Hongbo Huang, Weijun Chen, Yongxin Su, Yahui Liu, Xiuying Wang
2022 arXiv   pre-print
These algorithms can be generally divided into two categories, i.e. two-stage detector like Faster R-CNN and one-stage detector like YOLO.  ...  In recent years, face detection algorithms based on deep learning have made great progress.  ...  Occlusion-Aware Repulsion Loss Intra class occlusion may cause face A contains the features of face B, resulting in a higher false detection rate.  ... 
arXiv:2208.02019v2 fatcat:s6xeew7aynghjcwkkduiwyoy3i

Illumination-aware Faster R-CNN for Robust Multispectral Pedestrian Detection [article]

Chengyang Li, Dan Song, Ruofeng Tong, Min Tang
2018 arXiv   pre-print
With this in mind, we propose an Illumination-aware Faster R-CNN (IAF RCNN). Specifically, an Illumination-aware Network is introduced to give an illumination measure of the input image.  ...  The experimental results on KAIST Multispectral Pedestrian Benchmark validate the effectiveness of the proposed IAF R-CNN.  ...  Min Tang is supported in part by NSFC (61572423,61732015) and Zhejiang Provincial NSFC (LZ16F020003).  ... 
arXiv:1803.05347v2 fatcat:vmaeoqky5ncsbcngbdfeyrfu2m

Multi-Scale and Occlusion Aware Network for Vehicle Detection and Segmentation on UAV Aerial Images

Wang Zhang, Chunsheng Liu, Faliang Chang, Ye Song
2020 Remote Sensing  
In seeking to address these challenges, we propose a novel Multi-Scale and Occlusion Aware Network (MSOA-Net) for UAV based vehicle segmentation, which consists of two parts including a Multi-Scale Feature  ...  In this study, we release a large comprehensive UAV based vehicle segmentation dataset (UVSD), which is the first public dataset for UAV based vehicle detection and segmentation.  ...  Mask R-CNN [19] follows the idea of a two-stage object detection method, and adds a mask prediction branch on the basis of Faster R-CNN [5] .  ... 
doi:10.3390/rs12111760 fatcat:ft6uc2lxarbojlz454piflewpi

Improving Faster R-CNN Framework for Fast Vehicle Detection

Hoanh Nguyen
2019 Mathematical Problems in Engineering  
This paper proposes an improved framework based on Faster R-CNN for fast vehicle detection. Firstly, MobileNet architecture is adopted to build the base convolution layer in Faster R-CNN.  ...  However, due to large vehicle scale variation, heavy occlusion, or truncation of the vehicle in an image, recent deep CNN-based object detectors still showed a limited performance.  ...  R-CNN module, and the part-aware NMS is proposed to refine final results.  ... 
doi:10.1155/2019/3808064 fatcat:3nqjq3czejg2naasgry4vtlaau

Self-Enhanced R-CNNs for Human Detection with Semi-Supervised Assumptions

Xuexian Chen, Si Wu, Zhiwen Yu.
2020 IEEE Access  
To reduce the reliance of detection models on large amount of labeled data, we modify Faster R-CNN to facilitate semi-supervised human detection.  ...  Specifically, a Reliability Analysis (RA) module is included as an add-on into our Self-Enhanced R-CNN (SE-RCNN) model.  ...  They also investigated the effectiveness of different strategies in improving the performance of Faster R-CNN in [34] . To tackle occlusion in crowd scenes, Wang et al.  ... 
doi:10.1109/access.2020.2967414 fatcat:sxukl2ygmnehvb4ye63n2ndjo4

PFF-CB: Multiscale Occlusion Pedestrian Detection Method Based on PFF and CBAM

Guiyi Yang, Zhengyou Wang, Shanna Zhuang, Hui Wang, Dalin Zhang
2022 Computational Intelligence and Neuroscience  
However, in the scene of crowd occlusion or severe pedestrian occlusion, only small parts of the body can be used for detection.  ...  The results show that the PFF-CB module makes a good performance in occlusion pedestrian detection tasks.  ...  R-CNN to improve the detection accuracy in the crowd.  ... 
doi:10.1155/2022/3798060 pmid:35498206 pmcid:PMC9050291 fatcat:2cnycrznb5g2rktirjulx7ls7q

PANDA: A Gigapixel-Level Human-Centric Video Dataset

Xueyang Wang, Xiya Zhang, Yinheng Zhu, Yuchen Guo, Xiaoyun Yuan, Liuyu Xiang, Zerun Wang, Guiguang Ding, David Brady, Qionghai Dai, Lu Fang
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
A representative video Marathon of PANDA dataset. The characteristic of joint wide field-of-view and high spatial resolution empowers the large-scale, long-term, and multi-object visual analysis.  ...  As shown in Tab. 3, Faster R-CNN, Cascade R-CNN and RetinaNet show the difficulty in detecting small objects, resulting in very low precision and recall.  ...  Table 3 . 3 Performance of detection methods on PANDA. FR, CR, and RN denote Faster R-CNN, Cascade R-CNN and RetinaNet respectively.  ... 
doi:10.1109/cvpr42600.2020.00333 dblp:conf/cvpr/WangZZGYXWDBDF20 fatcat:sahedsewqjc5zfg3sfshvzseh4

PedHunter: Occlusion Robust Pedestrian Detector in Crowded Scenes [article]

Cheng Chi, Shifeng Zhang, Junliang Xing, Zhen Lei, Stan Z. Li, Xudong Zou
2019 arXiv   pre-print
Pedestrian detection in crowded scenes is a challenging problem, because occlusion happens frequently among different pedestrians.  ...  In this paper, we propose an effective and efficient detection network to hunt pedestrians in crowd scenes.  ...  During inference, even a pedestrian does not have a visible head, the learned better features for this pedestrian can also help the R-CNN branch for better detection performance.  ... 
arXiv:1909.06826v1 fatcat:bddgvqignngrlos4l6mbluavka

PedHunter: Occlusion Robust Pedestrian Detector in Crowded Scenes

Cheng Chi, Shifeng Zhang, Junliang Xing, Zhen Lei, Stan Z. Li, Xudong Zou
2020 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
Pedestrian detection in crowded scenes is a challenging problem, because occlusion happens frequently among different pedestrians.  ...  In this paper, we propose an effective and efficient detection network to hunt pedestrians in crowd scenes.  ...  the presented occlusion-simulated data augmentation for occluded pedestrian detection in a crowd.  ... 
doi:10.1609/aaai.v34i07.6690 fatcat:5m6uw4exn5asbfxwjyktf5df74

WiderPerson: A Diverse Dataset for Dense Pedestrian Detection in the Wild [article]

Shifeng Zhang, Yiliang Xie, Jun Wan, Hansheng Xia, Stan Z. Li, Guodong Guo
2019 arXiv   pre-print
We introduce an improved Faster R-CNN and the vanilla RetinaNet to serve as baselines for the new pedestrian detection benchmark.  ...  To narrow this gap and facilitate future pedestrian detection research, we introduce a large and diverse dataset named WiderPerson for dense pedestrian detection in the wild.  ...  Our improved Faster R-CNN achieves 5.49% M R for pedestrians on the (a) False positive: a background region is mistakenly detected as a pedestrian, or a pedestrian is detected with a misaligned bounding  ... 
arXiv:1909.12118v1 fatcat:heavnreizzhvtg5qiuzhen7kbq

From Handcrafted to Deep Features for Pedestrian Detection: A Survey [article]

Jiale Cao, Yanwei Pang, Jin Xie, Fahad Shahbaz Khan, Ling Shao
2020 arXiv   pre-print
Here we present a comprehensive survey on recent advances in pedestrian detection.  ...  Pedestrian detection is an important but challenging problem in computer vision, especially in human-centric tasks.  ...  In contrast, Intraclass occlusion occurs when pedestrians are occluded by other pedestrians. Intra-class occlusion is also called crowd occlusion [169] .  ... 
arXiv:2010.00456v1 fatcat:o2n7tpammzeyra2sl33hc5gtli

PS-RCNN: Detecting Secondary Human Instances in a Crowd via Primary Object Suppression [article]

Zheng Ge, Zequn Jie, Xin Huang, Rong Xu, Osamu Yoshie
2020 arXiv   pre-print
Detecting human bodies in highly crowded scenes is a challenging problem.  ...  After that, PS-RCNN utilizes another R-CNN module specialized in heavily occluded human detection (referred as S-RCNN) to detect the rest missed objects by P-RCNN.  ...  Crowd occlusion (intra-class occlusion and inter-class occlusion) is the main challenge in pedestrian detection. [2] and [3] propose two crowd human datasets (i.e.  ... 
arXiv:2003.07080v1 fatcat:63dtsyyisreovat7rqyacolse4
« Previous Showing results 1 — 15 out of 310 results