Skip to content

Latest commit

 

History

History
 
 

yolox

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

YOLOX

YOLOX: Exceeding YOLO Series in 2021

Abstract

In this report, we present some experienced improvements to YOLO series, forming a new high-performance detector -- YOLOX. We switch the YOLO detector to an anchor-free manner and conduct other advanced detection techniques, i.e., a decoupled head and the leading label assignment strategy SimOTA to achieve state-of-the-art results across a large scale range of models: For YOLO-Nano with only 0.91M parameters and 1.08G FLOPs, we get 25.3% AP on COCO, surpassing NanoDet by 1.8% AP; for YOLOv3, one of the most widely used detectors in industry, we boost it to 47.3% AP on COCO, outperforming the current best practice by 3.0% AP; for YOLOX-L with roughly the same amount of parameters as YOLOv4-CSP, YOLOv5-L, we achieve 50.0% AP on COCO at a speed of 68.9 FPS on Tesla V100, exceeding YOLOv5-L by 1.8% AP. Further, we won the 1st Place on Streaming Perception Challenge (Workshop on Autonomous Driving at CVPR 2021) using a single YOLOX-L model. We hope this report can provide useful experience for developers and researchers in practical scenes, and we also provide deploy versions with ONNX, TensorRT, NCNN, and Openvino supported.

YOLOX-l model structure

🥳 🚀 Results and Models

Backbone Size Batch Size AMP RTMDet-Hyp Mem (GB) Box AP Config Download
YOLOX-tiny 416 8xb8 No No 2.8 32.7 config model | log
YOLOX-tiny 416 8xb32 Yes Yes 4.9 34.3 (+1.6) config model | log
YOLOX-s 640 8xb8 Yes No 2.9 40.7 config model | log
YOLOX-s 640 8xb32 Yes Yes 9.8 41.9 (+1.2) config model | log
YOLOX-m 640 8xb8 Yes No 4.9 46.9 config model | log
YOLOX-m 640 8xb32 Yes Yes 17.6 47.5 (+0.6) config model | log
YOLOX-l 640 8xb8 Yes No 8.0 50.1 config model | log
YOLOX-x 640 8xb8 Yes No 9.8 51.4 config model | log

YOLOX uses a default training configuration of 8xbs8 which results in a long training time, we expect it to use 8xbs32 to speed up the training and not cause a decrease in mAP. We modified train_batch_size_per_gpu from 8 to 32, batch_augments_interval from 10 to 1 and base_lr from 0.01 to 0.04 under YOLOX-s default configuration based on the linear scaling rule, which resulted in mAP degradation. Finally, I found that using RTMDet's training hyperparameter can improve performance in YOLOX Tiny/S/M, which also validates the superiority of RTMDet's training hyperparameter.

The modified training parameters are as follows:

  1. train_batch_size_per_gpu: 8 -> 32
  2. batch_augments_interval: 10 -> 1
  3. num_last_epochs: 15 -> 20
  4. optim cfg: SGD -> AdamW, base_lr 0.01 -> 0.004, weight_decay 0.0005 -> 0.05
  5. ema momentum: 0.0001 -> 0.0002

Note:

  1. The test score threshold is 0.001.
  2. Due to the need for pre-training weights, we cannot reproduce the performance of the yolox-nano model. Please refer to Megvii-BaseDetection/YOLOX#674 for more information.

Citation

@article{yolox2021,
  title={{YOLOX}: Exceeding YOLO Series in 2021},
  author={Ge, Zheng and Liu, Songtao and Wang, Feng and Li, Zeming and Sun, Jian},
  journal={arXiv preprint arXiv:2107.08430},
  year={2021}
}