CE-FPN: Enhancing Channel Information for Object Detection

Luo, Yi; Zhang, Juntao; Cao, Xiang; Guo, Jingjuan; Shen, Haibo; Wang, Tianjiang; Qi, Feng

doi:10.48550/arxiv.2103.10643

Cited by 7 publications

(11 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…YOLOv3-R2 introduces Four AF blocks (FAF) into YOLOv3-R1. YOLOv3-R3 and YOLOv3-R4 introduces the CE-FPN [19] and BIFPN [20], two newly reported improvements to FPN, into YOLOv-R2 for achieving multiscale respectively. As shown in Table 2, anchor clustering by K-medians can increase mAP by 0.99% and FPS by 15.36%; FAF can further improve mAP by 4.82% while decrease FPS by 15.05%; the twin-tower structure can further increase mAP by 0.90% and decrease FPS by 8.17%.…”

Section: Methodsmentioning

confidence: 99%

MSFF: A Multi-Scale Feature Fusion Network for Surface Defect Detection of Aluminum Profiles

Sun

WEI

et al. 2022

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

This paper presents an improved YOLOv3 network, named MSFF-YOLOv3, for precisely detecting variable surface defects of aluminum profiles in practice. First, we introduce a larger prediction scale to provide detailed information for small defect detection; second, we design an efficient attention-guided block to extract more features of defects with less overhead; third, we design a bottom-up pyramid and integrate it with the existing feature pyramid network to construct a twin-tower structure to improve the circulation and fusion of features of different layers. In addition, we employ the K-median algorithm for anchor clustering to speed up the network reasoning. Experimental results showed that the mean average precision of the proposed network MSFF-YOLOv3 is higher than all conventional networks for surface defect detection of aluminum profiles. Moreover, the number of frames processed per second for our proposed MSFF-YOLOv3 could meet real-time requirements.

show abstract

Section: Methodsmentioning

confidence: 99%

MSFF: A Multi-Scale Feature Fusion Network for Surface Defect Detection of Aluminum Profiles

Sun

WEI

et al. 2022

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

show abstract

“…Yet recently, it has been found that the feature pyramid networks can cause serious aliasing effects in the process of fusing feature maps. To solve this problem, Luo et al [11] proposed an object detection method with enhanced channel information, which introduces a channel attention module to eliminate aliasing. In addition, they also proposed a subpixel jump fusion method, which effectively reduces the information loss in the process of channel reduction.…”

Section: Small Objects Detection Based On Multi-scale Feature Extractionmentioning

confidence: 99%

“…The existing small objects detection algorithms can effectively address the above problems. Such methods can be roughly divided into three categories: small objects detection methods based on multi-scale feature extraction [7][8][9][10][11], high-resolution feature-assisted small objects detection methods [12][13][14][15][16] and small objects detection methods guided by content information [17][18][19][20][21]. These methods focus on how to effectively improve the representational capability of the model to extract the discriminative features of small objects.…”

Section: Introductionmentioning

confidence: 99%

Texture and semantic integrated small objects detection in foggy scenes

Fang

Wang

et al. 2022

PLoS ONE

View full text Add to dashboard Cite

In recent years, small objects detection has received extensive attention from scholars for its important value in application. Some effective methods for small objects detection have been proposed. However, the data collected in real scenes are often foggy images, so the models trained with these methods are difficult to extract discriminative object features from such images. In addition, the existing small objects detection algorithms ignore the texture information and high-level semantic information of tiny objects, which limits the improvement of detection performance. Aiming at the above problems, this paper proposes a texture and semantic integrated small objects detection in foggy scenes. The algorithm focuses on extracting discriminative features unaffected by the environment, and obtaining texture information and high-level semantic information of small objects. Specifically, considering the adverse impact of foggy images on recognition performance, a knowledge guidance module is designed, and the discriminative features extracted from clear images by the model are used to guide the network to learn foggy images. Second, the features of high-resolution images and low-resolution images are extracted, and the adversarial learning method is adopted to train the model to give the network the ability to obtain the texture information of tiny objects from low-resolution images. Finally, an attention mechanism is constructed between feature maps of the same scale and different scales to further enrich the high-level semantic information of small objects. A large number of experiments have been conducted on data sets such as “Cityscape to Foggy” and “CoCo”. The mean prediction accuracy (mAP) has reached 46.2% on “Cityscape to Fogg”, and 33.3% on “CoCo”, which fully proves the effectiveness and superiority of the proposed method.

show abstract

“…Therefore, we consider using the dynamic scheme mentioned in Dynamic R-CNN (Zhang et al, 2020) As a result, when evaluating MT-FPN, by replacing FPN with MT-FPN, Faster R-CNN and Libra R-CNN obtain 42.5 and 42.4 Average Precision (AP) when using ResNet-50, respectively. When using ResNet-101 (He et al, 2016), Faster R-CNN based on MT-FPN obtains 43.6 AP, which surpasses the gains brought by the most advanced FPN-based methods, such as Libra R-CNN (Pang et al, 2019) and AugFPN (Guo et al, 2020) and CE-FPN (Luo et al, 2021). When evaluating the effect of the combined application of MT-FPN and DBLL, Faster R-CNN uses ResNet-50 and ResNext101-64x4d (Xie et al, 2017) as the backbone.…”

Section: Introductionmentioning

confidence: 99%

Object detection based on multiple trick feature pyramid networks and dynamic balanced L1 loss

Yin¹,

Zhang²

2022

IJWMC

View full text Add to dashboard Cite

Although the performance of the object detection has been significantly optimised in recent years, there is still a lot of room for designing multi-scale feature fusion methods and designing loss functions. Specifically, we propose Multiple Trick Feature Pyramid Networks (MT-FPN), by using various techniques such as feedback information, global module, attention mechanism, and fusion of refined information, to solve the problem of insufficient multi-scale feature fusion. We also propose Dynamic Balanced L1 Loss (DBLL), by utilising dynamic strategies and solving the derivative discontinuity problem, in order to help relieve the inconsistent problem between the dynamic training process and the fixed parameters. Moreover, by replacing FPN with MT-FPN, our Average Precision (AP) on Microsoft Common Objects in Context (MSCOCO) is 5.1 points and 3.8 points higher than FPN Faster R-CNN and Libra R-CNN, respectively. Without any bells and whistles, our experiments also show that the combined application of MT-FPN and DBLL achieves competitive performance compared with most advanced detectors on MS COCO benchmark.

show abstract

CE-FPN: Enhancing Channel Information for Object Detection

Cited by 7 publications

References 0 publications

MSFF: A Multi-Scale Feature Fusion Network for Surface Defect Detection of Aluminum Profiles

MSFF: A Multi-Scale Feature Fusion Network for Surface Defect Detection of Aluminum Profiles

Texture and semantic integrated small objects detection in foggy scenes

Object detection based on multiple trick feature pyramid networks and dynamic balanced L1 loss

Contact Info

Product

Resources

About