YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors

Wang, Chien-Yao; Bochkovskiy, Alexey; Liao, Hong-Yuan Mark

doi:10.48550/arxiv.2207.02696

Cited by 1,028 publications

(1,311 citation statements)

References 69 publications

Supporting

Mentioning

1,291

Contrasting

Unclassified

Order By: Relevance

“…We found that an F1 ∼ 0.7 can be achieved at precision 70% even with an extremely limited training set, establishing This work demonstrates that modern region-based object detection architectures -i.e., architectures that identify and then classify regions of interest as opposed to sliding window, classification-only architectures that classify every sub-region of an image -produce detection accuracies for LEs that allow for the searching of extremely large upcoming data sets for these objects, a task which may very well be unfeasible with slidingwindow classification models. And while it is true that these region-based models are rapidly evolving (YOLO itself is now at version 7, Wang et al 2022, and other more recent models include Chen et al 2019;Wu et al 2019 for example), the core conceptual foundations are common to all such models. That is, the models pre-sented in this paper have demonstrated that regionbased models are applicable to astronomical data (low SNR, very high spatial resolution, etc.)…”

Section: Discussionmentioning

confidence: 99%

Toward automated detection of light echoes in synoptic surveys: considerations on the application of the Deep Convolutional Neural Networks

Li¹,

Bianco²,

Dobler³

et al. 2022

Preprint

View full text Add to dashboard Cite

Light Echoes (LEs) are the reflections of astrophysical transients off of interstellar dust. They are fascinating astronomical phenomena that enable studies of the scattering dust as well as of the original transients. LEs, however, are rare and extremely difficult to detect as they appear as faint, diffuse, time-evolving features. The detection of LEs still largely relies on human inspection of images, a method unfeasible in the era of large synoptic surveys. The Vera C. Rubin Observatory Legacy Survey of Space and Time, LSST, will generate an unprecedented amount of astronomical imaging data at high spatial resolution, exquisite image quality, and over tens of thousands of square degrees of sky: an ideal survey for LEs. However, the Rubin data processing pipelines are optimized for the detection of point-sources and will entirely miss LEs. Over the past several years, Artificial Intelligence (AI) object detection frameworks have achieved and surpassed real-time, human-level performance. In this work, we prepare a dataset from the ATLAS telescope and test a popular AI object detection framework, You Only Look Once, or YOLO, developed in the computer vision community, to demonstrate the potential of AI in the detection of LEs in astronomical images. We find that an AI framework can reach human-level performance even with a size-and quality-limited dataset. We explore and highlight challenges, including class imbalance and label incompleteness, and roadmap the work required to build an end-to-end pipeline for the automated detection and study of LEs in high-throughput astronomical surveys.

show abstract

Section: Discussionmentioning

confidence: 99%

Toward automated detection of light echoes in synoptic surveys: considerations on the application of the Deep Convolutional Neural Networks

Li¹,

Bianco²,

Dobler³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Among them, YOLOv5n is the latest lightweight algorithm of YOLOv5, and YOLOv3 is a relatively mature large-scale one-stage detection algorithm. YOLOv7 [ 28 ] is the latest algorithm of YOLO family at present, which has the strongest comprehensive performance in full-scale detection, and YOLOv7-tiny is a lightweight version of YOLOv7, which has similar parameter quantities and calculation quantities with SF-YOLOv5. ResNeXt-CSP is a new detector combined with classical algorithms ResNeXt [ 29 ] and CSPNet [ 32 ], which have excellent performance.…”

Section: Methodsmentioning

confidence: 99%

“…have been introduced. In addition, the latest YOLOv7 [ 28 ] (which is still under update) and various improved algorithms based on ResNet [ 29 , 30 ] also have excellent performance.…”

Section: Relevant Workmentioning

confidence: 99%

SF-YOLOv5: A Lightweight Small Object Detection Algorithm Based on Improved Feature Fusion Mode

Liu

Sun

Deng

2022

Sensors

101

View full text Add to dashboard Cite

In the research of computer vision, a very challenging problem is the detection of small objects. The existing detection algorithms often focus on detecting full-scale objects, without making proprietary optimization for detecting small-size objects. For small objects dense scenes, not only the accuracy is low, but also there is a certain waste of computing resources. An improved detection algorithm was proposed for small objects based on YOLOv5. By reasonably clipping the feature map output of the large object detection layer, the computing resources required by the model were significantly reduced and the model becomes more lightweight. An improved feature fusion method (PB-FPN) for small object detection based on PANet and BiFPN was proposed, which effectively increased the detection ability for small object of the algorithm. By introducing the spatial pyramid pooling (SPP) in the backbone network into the feature fusion network and connecting with the model prediction head, the performance of the algorithm was effectively enhanced. The experiments demonstrated that the improved algorithm has very good results in detection accuracy and real-time ability. Compared with the classical YOLOv5, the mAP@0.5 and mAP@0.5:0.95 of SF-YOLOv5 were increased by 1.6% and 0.8%, respectively, the number of parameters of the network were reduced by 68.2%, computational resources (FLOPs) were reduced by 12.7%, and the inferring time of the mode was reduced by 6.9%.

show abstract

“…Redmon [ 6 ] et al, proposed the YOLO object detection framework, a neural network framework appropriate for targets detected in real-time and with a detection speed of 45 FPS for the first time. Later versions of YOLO, YOLO 9000, YOLO v3, YOLOv4, and YOLOv7 [ 7 , 8 , 9 , 10 ] included new neural network architectures such as batch normalization [ 11 ], FPN [ 12 ], and SPP [ 13 ] to strike a balance between detection accuracy and speed. To build a genuinely anchorless detector, Tian [ 14 ] et al, suggested Fully Convolutional One-Stage Object Detection (FCOS) based on the RetinaNet [ 15 ] network architecture.…”

Section: Introductionmentioning

confidence: 99%

Multi-Object Detection in Security Screening Scene Based on Convolutional Neural Network

Sun

Zhang

Liu

et al. 2022

Sensors

View full text Add to dashboard Cite

The technique for target detection based on a convolutional neural network has been widely implemented in the industry. However, the detection accuracy of X-ray images in security screening scenarios still requires improvement. This paper proposes a coupled multi-scale feature extraction and multi-scale attention architecture. We integrate this architecture into the Single Shot MultiBox Detector (SSD) algorithm and find that it can significantly improve the effectiveness of target detection. Firstly, ResNet is used as the backbone network to replace the original VGG network to improve the feature extraction capability of the convolutional neural network for images. Secondly, a multi-scale feature extraction (MSE) structure is designed to enrich the information contained in the multi-stage prediction feature layer. Finally, the multi-scale attention architecture (MSA) is fused onto the prediction feature layer to eliminate the redundant features’ interference and extract effective contextual information. In addition, a combination of Adaptive-NMS and Soft-NMS is used to output the final prediction anchor boxes when performing non-maximum suppression. The results of the experiments show that the improved method improves the mean average precision (mAP) value by 7.4% compared to the original approach. New modules make detection much more accurate while keeping the detection speed the same.

show abstract

YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors

Cited by 1,028 publications

References 69 publications

Toward automated detection of light echoes in synoptic surveys: considerations on the application of the Deep Convolutional Neural Networks

Toward automated detection of light echoes in synoptic surveys: considerations on the application of the Deep Convolutional Neural Networks

SF-YOLOv5: A Lightweight Small Object Detection Algorithm Based on Improved Feature Fusion Mode

Multi-Object Detection in Security Screening Scene Based on Convolutional Neural Network

Contact Info

Product

Resources

About