YOLOv6: A Single-Stage Object Detection Framework for Industrial Applications
By Chuyin Li, Lu Li, Hongliang Jiang, Kaiheng Weng, Yifei Geng, Liang Li, Zaidan Ke, Qingyuan Li, Meng Cheng, Weiqiang Nie and 7 more. arXiv.org, 2022
3,214 citations recorded by the source corpus.
For years, the YOLO series has been the de facto industry-level standard for efficient object detection. The YOLO community has prospered overwhelmingly to enrich its use in a multitude of hardware platforms and abundant scenarios. In this technical report, we strive to push its limits to the next level, stepping forward with an unwavering mindset for industry application. Considering the diverse requirements for speed and accuracy in the real environment, we extensively examine the up-to-date object detection advancements either from industry or academia. Specifically, we heavily assimilate ideas from recent network design, training strategies, testing techniques, quantization, and optimization methods. On top of this, we integrate our thoughts and practice to build a suite of deployment-ready networks at various scales to accommodate diversified use cases. With the generous permission of YOLO authors, we name it YOLOv6. We also express our warm welcome to users and contributors for further enhancement. For a glimpse of performance, our YOLOv6-N hits 35.9% AP on the COCO dataset at a throughput of 1234 FPS on an NVIDIA Tesla T4 GPU. YOLOv6-S strikes 43.5% AP at 495 FPS, outperforming other mainstream detectors at the same scale~(YOLOv5-S, YOLOX-S, and PPYOLOE-S). Our quantized version of YOLOv6-S even brings a new state-of-the-art 43.3% AP at 869 FPS. Furthermore, YOLOv6-M/L also achieves better accuracy performance (i.e., 49.5%/52.3%) than other detectors with a similar inference speed. We carefully conducted experiments to validate the effectiveness of each component. Our code is made available at https://github.com/meituan/YOLOv6.
Subfields: Computer Vision and Pattern Recognition, Industrial and Manufacturing Engineering
View Paper: https://www.semanticscholar.org/paper/ee19e8936275c8efe30c91a3c64e5ad8f3b15dc3
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computer Vision and Pattern Recognition
- Hongpeng Liao
- Chien-Yao Wang
- Diana-Margarita Córdova-Esparza
- Juan R. Terven
- Yunhe Wang
- Xinggang Wang
- I-Hau Yeh
- Kai Han
In Industrial and Manufacturing Engineering
- Hongpeng Liao
- Chien-Yao Wang
- I-Hau Yeh
- Diana-Margarita Córdova-Esparza
- Kai Chen
- Muhammad Hussain
- Xinggang Wang
- Juan R. Terven
Citations in the corpus
- YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information (2024, methodology): “The current mainstream real-time object detectors are the YOLO series [2, 7, 13–15, 25, 30, 31, 47–49, 61–63, 74, 75], and most of these models use CSPNet [64] or ELAN [65] and their variants as the main computing units.”
- YOLO-based Object Detection Models: A Review and its Applications (2024, neutral): “YOLOv6 [82] was published by Meituan Incorporates in September 2022.”
- YOLOv6 v3.0: A Full-Scale Reloading (2023, methodology): “The evaluation is made consistent with the early versions of YOLOv6 [7], which focuses on the throughput and the GPU latency at deployment.”
- Gold-YOLO: Efficient Object Detector via Gather-and-Distribute Mechanism (2023, methodology): “Small-scale object detection models based on CNN still dominate the speed-accuracy trade-off, such as YOLOX [10] and YOLOv6-v8 [20, 35, 12].”
- YOLO-World: Real-Time Open-Vocabulary Object Detection (2024, background): “Recent works [9, 14, 48, 53, 59] have explored the prevalent vision-language models [19, 39] to address open-20”
- YOLOv1 to v8: Unveiling Each Variant–A Comprehensive Review of YOLO (2024, background): “Meituan Vision AI Department [41] introduced YOLOv6 in September 2022 [42], boasting several innovative features claiming to enhance both efficiency and accuracy.”
- Hyper-YOLO: When Visual Object Detection Meets Hypergraph Computation (2024, neutral): “T HE YOLO series [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11] stands out as a mainstream method in the realm of object detection, offering several advantages that cater to these diverse applications.”
- RTMDet: An Empirical Study of Designing Real-Time Object Detectors (2022, methodology): “Cross-sample augmentations such as MixUp [94] and CutMix [93] are widely adopted in recent object detectors [3, 21, 25, 42, 71].”
- DAMO-YOLO : A Report on Real-Time Object Detection Design (2022, methodology): “Following YOLO-Series [9, 18, 32, 34] model exponential moving average (EMA) and grouped weight decay are used.”
- A Comprehensive Review of YOLO Architectures in Computer Vision: From YOLOv1 to YOLOv8 and YOLO-NAS (2023, methodology): “YOLOv6 [97] was published in ArXiv in September 2022 by Meituan Vision AI Department.”
- A Comprehensive Review of YOLO: From YOLOv1 to YOLOv8 and Beyond (2023, methodology): “YOLOv6 [86] was published in ArXiv in September 2022 by Meituan Vision AI Department.”
- A review of convolutional neural networks in computer vision (2024, positive): “YOLOv6 (Li et al. 2022), designed for industrial applications, was released in 2022, and it supports the entire chain of industrial application requirements, such as model training, inference, and multi-platform deployment, as well as making several improvements and optimizations at the network…”