Disentangling classification and regression in Siamese‐based network for visual tracking

Section: ) End-to-end Results From All Servicesmentioning

confidence: 99%

“…More examples of using multiple layers from the FPN, Zhang et al [23] used multiple layers of FPN. Utilizing multiple layers in this study allowed the network to acquire hierarchical representations.…”

Section: B Multiple Layers-based Detectormentioning

confidence: 99%

Efficient Text Bounding Box Identification Using Mask R-CNN: Case of Thai Documents

Kiatphaisansophon,

Wanvarie,

Cooharojananone

2024

“…Reference [42] proposes an active pedestrian detector to address the performance degradation problem of smallsized pedestrian detection by leveraging the rich feature hierarchy and initial pedestrian proposals provided by ResNet and Faster R-CNN, which operate explicitly on multilayer neural representations, achieving significant false detection rate reduction. Reference [43] An effective visual tracking framework called SiamDTH is proposed, which distinguishes in Siamese-based network-based visual tracking through feature decoupling and different tracking head structures. Classification and regression tasks, thereby obtaining more robust classification predictions and accurate regression predictions.…”

Section: E Pedestrian Detectionmentioning

confidence: 99%

Small-Scale Pedestrian Detection Using Fusion Network and Probabilistic Loss

Zhang,

Yang,

Liu

et al. 2024

Small-scale pedestrian detection is a challenge. The main issues are as follows: (1) Troubled by their small scale, it is difficult to extract features effectively; (2) During the detection process, it is easily disturbed by background noise such as inter-class occlusion and intra-class occlusion, leading to missed or false detection; (3) The current widely used IoU measurement method is very sensitive to the position deviation of small objects, which seriously reduces the detection performance. To address these problems, we improve YOLOv5 structure by integrating Non-Local and Convolution structures, building a new feature extraction module called ResNet-Conv&NonL, combined with the ResNet structure. This module was then integrated into the backbone of YOLOv5 for better image feature extraction. In addition, we developed a novel model to measure the similarity between bounding boxes, which are embedded in the loss function of the YOLOv5 structure to replace the normal IoU measurement. Experiments on a self-made dataset and a combined dataset from Caltech and CityPersons show the feasibility of the proposed network structure. Results demonstrate the feasibility of the improved network structure is superior to the original method because it increases average precision by 6.0% compared to the original one.

“…O BJECT detection [1], as a longstanding, fundamental and challenging problem in computer vision [2], has been an active field of research for several decades [3], [4]. The task of object detection is to identify object categories and predict the location of each object in an image by a bounding box, and there are many real world applications [5] based on this task, such as face detection and pedestrian detection [6].…”

Section: Introductionmentioning

confidence: 99%

MSF-CSPNet: A Specially Designed Backbone Network for Faster R-CNN

Li,

Sun,

Dong

et al. 2024

Although Faster R-CNN has undergone a lot of improvements, it still exists a significant gap in the performance between the detection of small and large objects, mainly because the low-level network lacks semantic information and small objects are only involved in a few images. To mitigate the above issues, we propose an object detection model based on Multi-Scale Feature fusion Cross Stage Partial Network (MSF-CSPNet) in this paper. The proposed MSF-CSPNet focuses on the fusion of concrete features and abstract features from multi-scale feature by learning shallow features at the shallow level and deep features at the deep level. Meanwhile, the data augmentation is performed by using random horizontal flip. On the basis, the improved Faster-RCNN model with Automatic Mixed Precision, Group Batch Sampler and MSF-CSPNet was formed. The proposed algorithm is valuated on the Microsoft Common Objects in Context (MS COCO) 2017 and obtained leading performance with 5.4% improvement in AP coco , 5.9% improvement in AP 50 , 6.9% improvement in AP 75 , 5.8% improvement in AP S , 6.1% improvement in AP M , 5.8% improvement in AP L compare to Faster R-CNN based on ResNet-50 with Feature Pyramid Network (FPN) backbone, and also outperformed previous reports on state-of-art Faster R-CNN series using other backbone networks, especially for small object detection. This research shows that the combination of a backbone with stronger learning ability and FPN is helpful to detect the expression of objects. Faster R-CNN based on MSF-CSPNet has high efficiency and better balance between accuracy and speed.