• Corpus ID: 233324262

PP-YOLOv2: A Practical Object Detector

  title={PP-YOLOv2: A Practical Object Detector},
  author={Xin Huang and Xinxin Wang and Wenyu Lv and Xiaying Bai and Xiang Long and Kaipeng Deng and Qingqing Dang and Shumin Han and Qiwen Liu and Xiaoguang Hu and Dianhai Yu and Yanjun Ma and Osamu Yoshie},
Being effective and efficient is essential to an object detector for practical use. To meet these two concerns, we comprehensively evaluate a collection of existing refinements to improve the performance of PP-YOLO while almost keep the infer time unchanged. This paper will analyze a collection of refinements and empirically evaluate their impact on the final model performance through incremental ablation study. Things we tried that didn’t work will also be discussed. By combining multiple… 

Figures and Tables from this paper

Medicinal Chrysanthemum Detection under Complex Environments Using the MC-LCNN Model
A novel lightweight convolutional neural network for medicinal chrysanthemum detection (MC-LCNN) is proposed and embedded into the edge computing device NVIDIA Jetson TX2 for real-time object detection, adopting a CPU–GPU multithreaded pipeline design to improve the inference speed by 2FPS.
2nd Place Solution for Waymo Open Dataset Challenge - Real-time 2D Object Detection
A real-time method to detect the 2D objects from images using several popular one-stage object detectors and train the models of variety input strategies independently, to yield better performance for accurate multi-scale detection of each category, especially for small objects.
YOLOSA: Object detection based on 2D local feature superimposed self-attention
This work proposes a novel self-attention module, called 2D local feature superimposed self-Attention, for the feature concatenation stage of the neck network, and proposes and optimize an efficient decoupled head and AB-OTA, and achieves SOTA results.
PCB defect detection based on PP-YOLOv2
Experimental results show that the deep learning-based target detection method proposed has high detection accuracy and fast detection speed, which is more suitable for production use compared with other PCB defect detection methods.
Pytri: A multi-weight detection system for biological entities
An online platform for biologists consisting of a system with multiple trained machine learning weights to detect various biological entities such as yeast colonies, bacterial colonies, and melanoma clusters, achieving significantly higher accuracy than traditional methods when compared to the base standard manual count.
Effective Multi-Frame Optical Detection Algorithm for GEO Space Objects
A deep-learning-based framework called PP-YOLOv2 for single-frame object detection is conducted and a post-processing algorithm named CFS is designed for further candidate filtration and supplement to obtain the eventual prediction results.
Semisupervised heterogeneous ensemble for ship target discrimination in synthetic aperture radar images
Ship detection using synthetic aperture radar (SAR) plays an important role in marine applications. The existing methods are capable of quickly obtaining many candidate targets, but numerous non-ship
Several YOLO-based methods are chosen for comparison and experiment results indicate that these methods can recognize the marine animals in coral reef quickly and accurately.
A Computer Vision-based System for Surgical Waste Detection
The world population is going through a difficult time due to the pandemic of COVID-19 while other disasters prevail. However, a new environmental catastrophe is coming because surgical masks and
PP-YOLOE: An evolved version of YOLO
PP-YOLOE, an industrial state-of-the-art object detector with high performance and friendly deployment is presented, using anchor-free paradigm, more pow-erful backbone and neck equipped with CSPRepResStage, ET-head and dynamic label assignment algorithm TAL.


Scaled-YOLOv4: Scaling Cross Stage Partial Network
We show that the YOLOv4 object detection neural network based on the CSP approach, scales both up and down and is applicable to small and large networks while maintaining optimal speed and accuracy.
YOLOv4: Optimal Speed and Accuracy of Object Detection
This work uses new features: WRC, CSP, CmBN, SAT, Mish activation, Mosaic data augmentation, C mBN, DropBlock regularization, and CIoU loss, and combine some of them to achieve state-of-the-art results: 43.5% AP for the MS COCO dataset at a realtime speed of ~65 FPS on Tesla V100.
Learning Spatial Fusion for Single-Shot Object Detection
This work proposes a novel and data driven strategy for pyramidal feature fusion, referred to as adaptively spatial feature fusion (ASFF), which learns the way to spatially filter conflictive information to suppress the inconsistency, thus improving the scale-invariance of features, and introduces nearly free inference overhead.
EfficientDet: Scalable and Efficient Object Detection
This paper systematically study neural network architecture design choices for object detection and proposes a weighted bi-directional feature pyramid network (BiFPN) and a compound scaling method that uniformly scales the resolution, depth, and width for all backbone, feature network, and box/class prediction networks at the same time.
Bag of Tricks for Image Classification with Convolutional Neural Networks
This paper examines a collection of training procedure refinements and empirically evaluates their impact on the final model accuracy through ablation study, and shows that by combining these refinements together, they are able to improve various CNN models significantly.
YOLOv3: An Incremental Improvement
We present some updates to YOLO! We made a bunch of little design changes to make it better. We also trained this new network that's pretty swell. It's a little bigger than last time but more
Path Aggregation Network for Instance Segmentation
Path Aggregation Network (PANet) is proposed aiming at boosting information flow in proposal-based instance segmentation framework by enhancing the entire feature hierarchy with accurate localization signals in lower layers by bottom-up path augmentation.
PP-YOLO: An Effective and Efficient Implementation of Object Detector
A new object detector based on YOLOv3 that can be directly applied in actual application scenarios, rather than propose a novel detection model, and can achieve a better balance between effectiveness and efficiency than existing state-of-the-art detectors such as EfficientDet and Y OLOv4.
Cascade R-CNN: Delving Into High Quality Object Detection
A simple implementation of the Cascade R-CNN is shown to surpass all single-model object detectors on the challenging COCO dataset, and experiments show that it is widely applicable across detector architectures, achieving consistent gains independently of the baseline detector strength.
mixup: Beyond Empirical Risk Minimization
This work proposes mixup, a simple learning principle that trains a neural network on convex combinations of pairs of examples and their labels, which improves the generalization of state-of-the-art neural network architectures.