add pphuman ppvehicle training configs and modelzoo (#6187)

* add pphuman ppvehicle configs and doc

* update doc and configs ,test=document_fix

* fix doc, test=document_fix

configs/pphuman/README.md

+简体中文 | [English](README.md)
+
+# PP-YOLOE Human 检测模型
+
+PaddleDetection团队提供了针对行人的基于PP-YOLOE的检测模型，用户可以下载模型进行使用。
+其中整理后的COCO格式的CrowdHuman数据集[下载链接](https://bj.bcebos.com/v1/paddledet/data/crowdhuman.zip)，检测类别仅一类 `pedestrian(1)`，原始数据集[下载链接](http://www.crowdhuman.org/download.html)。
+
+|    模型   |  数据集  | mAP<sup>val<br>0.5:0.95 | mAP<sup>val<br>0.5 |  下载  | 配置文件 |
+|:---------|:-------:|:------:|:------:| :----: | :------:|
+|PP-YOLOE-s|   CrowdHuman   |  42.5  |  77.9  | [下载链接](https://paddledet.bj.bcebos.com/models/ppyoloe_crn_s_36e_crowdhuman.pdparams) | [配置文件](./ppyoloe_crn_s_36e_crowdhuman.yml) |
+|PP-YOLOE-l|   CrowdHuman   |  48.0  |  81.9  | [下载链接](https://paddledet.bj.bcebos.com/models/ppyoloe_crn_l_36e_crowdhuman.pdparams) | [配置文件](./ppyoloe_crn_l_36e_crowdhuman.yml) |
+
+
+**注意:**
+- PP-YOLOE模型训练过程中使用8 GPUs进行混合精度训练，如果**GPU卡数**或者**batch size**发生了改变，你需要按照公式 **lr<sub>new</sub> = lr<sub>default</sub> * (batch_size<sub>new</sub> * GPU_number<sub>new</sub>) / (batch_size<sub>default</sub> * GPU_number<sub>default</sub>)** 调整学习率。
+- 具体使用教程请参考[ppyoloe](../ppyoloe#getting-start)。
+
+
+## 引用
+```
+@article{shao2018crowdhuman,
+    title={CrowdHuman: A Benchmark for Detecting Human in a Crowd},
+    author={Shao, Shuai and Zhao, Zijian and Li, Boxun and Xiao, Tete and Yu, Gang and Zhang, Xiangyu and Sun, Jian},
+    journal={arXiv preprint arXiv:1805.00123},
+    year={2018}
+  }
+```

configs/pphuman/ppyoloe_crn_l_36e_crowdhuman.yml

+_BASE_: [
+  '../datasets/coco_detection.yml',
+  '../runtime.yml',
+  '../ppyoloe/_base_/optimizer_300e.yml',
+  '../ppyoloe/_base_/ppyoloe_crn.yml',
+  '../ppyoloe/_base_/ppyoloe_reader.yml',
+]
+log_iter: 100
+snapshot_epoch: 4
+weights: output/ppyoloe_crn_l_36e_crowdhuman/model_final
+
+pretrain_weights: https://paddledet.bj.bcebos.com/models/ppyoloe_crn_l_300e_coco.pdparams
+depth_mult: 1.0
+width_mult: 1.0
+
+num_classes: 1
+TrainDataset:
+  !COCODataSet
+    image_dir: ""
+    anno_path: annotations/train.json
+    dataset_dir: dataset/crowdhuman
+    data_fields: ['image', 'gt_bbox', 'gt_class', 'is_crowd']
+
+EvalDataset:
+  !COCODataSet
+    image_dir: ""
+    anno_path: annotations/val.json
+    dataset_dir: dataset/crowdhuman
+
+TestDataset:
+  !ImageFolder
+    anno_path: annotations/val.json
+    dataset_dir: dataset/crowdhuman
+
+TrainReader:
+  batch_size: 8
+
+epoch: 36
+LearningRate:
+  base_lr: 0.001
+  schedulers:
+    - !CosineDecay
+      max_epochs: 43
+    - !LinearWarmup
+      start_factor: 0.
+      epochs: 1
+
+PPYOLOEHead:
+  static_assigner_epoch: -1
+  nms:
+    name: MultiClassNMS
+    nms_top_k: 1000
+    keep_top_k: 100
+    score_threshold: 0.01
+    nms_threshold: 0.6

configs/pphuman/ppyoloe_crn_s_36e_crowdhuman.yml

+_BASE_: [
+  '../datasets/coco_detection.yml',
+  '../runtime.yml',
+  '../ppyoloe/_base_/optimizer_300e.yml',
+  '../ppyoloe/_base_/ppyoloe_crn.yml',
+  '../ppyoloe/_base_/ppyoloe_reader.yml',
+]
+log_iter: 100
+snapshot_epoch: 4
+weights: output/ppyoloe_crn_s_36e_crowdhuman/model_final
+
+pretrain_weights: https://paddledet.bj.bcebos.com/models/ppyoloe_crn_s_300e_coco.pdparams
+depth_mult: 0.33
+width_mult: 0.50
+
+num_classes: 1
+TrainDataset:
+  !COCODataSet
+    image_dir: ""
+    anno_path: annotations/train.json
+    dataset_dir: dataset/crowdhuman
+    data_fields: ['image', 'gt_bbox', 'gt_class', 'is_crowd']
+
+EvalDataset:
+  !COCODataSet
+    image_dir: ""
+    anno_path: annotations/val.json
+    dataset_dir: dataset/crowdhuman
+
+TestDataset:
+  !ImageFolder
+    anno_path: annotations/val.json
+    dataset_dir: dataset/crowdhuman
+
+TrainReader:
+  batch_size: 8
+
+epoch: 36
+LearningRate:
+  base_lr: 0.001
+  schedulers:
+    - !CosineDecay
+      max_epochs: 43
+    - !LinearWarmup
+      start_factor: 0.
+      epochs: 1
+
+PPYOLOEHead:
+  static_assigner_epoch: -1
+  nms:
+    name: MultiClassNMS
+    nms_top_k: 1000
+    keep_top_k: 100
+    score_threshold: 0.01
+    nms_threshold: 0.6

configs/ppvehicle/README.md

+简体中文 | [English](README.md)
+
+# PP-YOLOE Vehicle 检测模型
+
+PaddleDetection团队提供了针对自动驾驶场景的基于PP-YOLOE的检测模型，用户可以下载模型进行使用，主要包含5个数据集(BDD100K-DET、BDD100K-MOT、UA-DETRAC、PPVehicle9cls、PPVehicle)。其中前3者为公开数据集，后两者为整合数据集。
+- BDD100K-DET具体类别为10类，包括`pedestrian(1), rider(2), car(3), truck(4), bus(5), train(6), motorcycle(7), bicycle(8), traffic light(9), traffic sign(10)`。
+- BDD100K-MOT具体类别为8类，包括`pedestrian(1), rider(2), car(3), truck(4), bus(5), train(6), motorcycle(7), bicycle(8)`，但数据集比BDD100K-DET更大更多。
+- UA-DETRAC具体类别为4类，包括`car(1), bus(2), van(3), others(4)`。
+- PPVehicle9cls数据集整合了BDD100K-MOT和UA-DETRAC，具体类别为9类，包括`pedestrian(1), rider(2), car(3), truck(4), bus(5), van(6), motorcycle(7), bicycle(8), others(9)`。
+- PPVehicle数据集整合了BDD100K-MOT和UA-DETRAC，是将BDD100K-MOT中的`car, truck, bus, van`和UA-DETRAC中的`car, bus, van`都合并为1类`vehicle(1)`后的数据集。
+
+
+|    模型   |       数据集     | 类别数  | mAP<sup>val<br>0.5:0.95 |  下载链接  | 配置文件 |
+|:---------|:---------------:|:------:|:-----------------------:|:---------:| :-----: |
+|PP-YOLOE-l|   BDD100K-DET   |   10   |  35.6 | [下载链接](https://paddledet.bj.bcebos.com/models/ppyoloe_crn_l_36e_bdd100kdet.pdparams) | [配置文件](./ppyoloe_crn_l_36e_bdd100kdet.yml) |
+|PP-YOLOE-l|   BDD100K-MOT   |   8    |  33.7 | [下载链接](https://paddledet.bj.bcebos.com/models/ppyoloe_crn_l_36e_bdd100kmot.pdparams) | [配置文件](./ppyoloe_crn_l_36e_bdd100kmot.yml) |
+|PP-YOLOE-l|   UA-DETRAC     |   4    |  51.4 | [下载链接](https://paddledet.bj.bcebos.com/models/ppyoloe_crn_l_36e_uadetrac.pdparams) | [配置文件](./ppyoloe_crn_l_36e_uadetrac.yml) |
+|PP-YOLOE-l|   PPVehicle9cls |   9    |  40.0 | [下载链接](https://paddledet.bj.bcebos.com/models/mot_ppyoloe_l_36e_ppvehicle9cls.pdparams) | [配置文件](./mot_ppyoloe_l_36e_ppvehicle9cls.yml) |
+|PP-YOLOE-s|   PPVehicle9cls |   9    |  35.3 | [下载链接](https://paddledet.bj.bcebos.com/models/mot_ppyoloe_s_36e_ppvehicle9cls.pdparams) | [配置文件](./mot_ppyoloe_s_36e_ppvehicle9cls.yml) |
+|PP-YOLOE-l|   PPVehicle     |   1    |  63.9 | [下载链接](https://paddledet.bj.bcebos.com/models/mot_ppyoloe_l_36e_ppvehicle.pdparams) | [配置文件](./mot_ppyoloe_l_36e_ppvehicle.yml) |
+|PP-YOLOE-s|   PPVehicle     |   1    |  61.3  | [下载链接](https://paddledet.bj.bcebos.com/models/mot_ppyoloe_s_36e_ppvehicle.pdparams) | [配置文件](./mot_ppyoloe_s_36e_ppvehicle.yml) |
+
+**注意:**
+- PP-YOLOE模型训练过程中使用8 GPUs进行混合精度训练，如果**GPU卡数**或者**batch size**发生了改变，你需要按照公式 **lr<sub>new</sub> = lr<sub>default</sub> * (batch_size<sub>new</sub> * GPU_number<sub>new</sub>) / (batch_size<sub>default</sub> * GPU_number<sub>default</sub>)** 调整学习率。
+- 具体使用教程请参考[ppyoloe](../ppyoloe#getting-start)。
+
+
+## 引用
+```
+@InProceedings{bdd100k,
+    author = {Yu, Fisher and Chen, Haofeng and Wang, Xin and Xian, Wenqi and Chen,
+              Yingying and Liu, Fangchen and Madhavan, Vashisht and Darrell, Trevor},
+    title = {BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learning},
+    booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
+    month = {June},
+    year = {2020}
+}
+
+@article{CVIU_UA-DETRAC,
+    author = {Longyin Wen and Dawei Du and Zhaowei Cai and Zhen Lei and Ming{-}Ching Chang and
+              Honggang Qi and Jongwoo Lim and Ming{-}Hsuan Yang and Siwei Lyu},
+    title = {{UA-DETRAC:} {A} New Benchmark and Protocol for Multi-Object Detection and Tracking},
+    journal = {Computer Vision and Image Understanding},
+    year = {2020}
+}  
+```

configs/ppvehicle/mot_ppyoloe_l_36e_ppvehicle.yml

+_BASE_: [
+  '../datasets/coco_detection.yml',
+  '../runtime.yml',
+  '../ppyoloe/_base_/optimizer_300e.yml',
+  '../ppyoloe/_base_/ppyoloe_crn.yml',
+  '../ppyoloe/_base_/ppyoloe_reader.yml',
+]
+log_iter: 100
+snapshot_epoch: 4
+weights: output/mot_ppyoloe_l_36e_ppvehicle/model_final
+
+pretrain_weights: https://paddledet.bj.bcebos.com/models/ppyoloe_crn_l_300e_coco.pdparams
+depth_mult: 1.0
+width_mult: 1.0
+
+num_classes: 1
+
+TrainDataset:
+  !COCODataSet
+    image_dir: ""
+    anno_path: annotations/train_all.json
+    dataset_dir: dataset/ppvehicle
+    data_fields: ['image', 'gt_bbox', 'gt_class', 'is_crowd']
+    allow_empty: true
+
+EvalDataset:
+  !COCODataSet
+    image_dir: ""
+    anno_path: annotations/val_all.json
+    dataset_dir: dataset/ppvehicle
+
+TestDataset:
+  !ImageFolder
+    anno_path: annotations/val_all.json
+    dataset_dir: dataset/ppvehicle
+
+TrainReader:
+  batch_size: 8
+
+epoch: 36
+LearningRate:
+  base_lr: 0.001
+  schedulers:
+    - !CosineDecay
+      max_epochs: 43
+    - !LinearWarmup
+      start_factor: 0.
+      epochs: 1
+
+PPYOLOEHead:
+  static_assigner_epoch: -1
+  nms:
+    name: MultiClassNMS
+    nms_top_k: 1000
+    keep_top_k: 100
+    score_threshold: 0.01
+    nms_threshold: 0.6

configs/ppvehicle/mot_ppyoloe_l_36e_ppvehicle9cls.yml

+_BASE_: [
+  '../datasets/coco_detection.yml',
+  '../runtime.yml',
+  '../ppyoloe/_base_/optimizer_300e.yml',
+  '../ppyoloe/_base_/ppyoloe_crn.yml',
+  '../ppyoloe/_base_/ppyoloe_reader.yml',
+]
+log_iter: 100
+snapshot_epoch: 4
+weights: output/mot_ppyoloe_l_36e_ppvehicle9cls/model_final
+
+pretrain_weights: https://paddledet.bj.bcebos.com/models/ppyoloe_crn_l_300e_coco.pdparams
+depth_mult: 1.0
+width_mult: 1.0
+
+num_classes: 9
+
+TrainDataset:
+  !COCODataSet
+    image_dir: ""
+    anno_path: annotations/train_all_9cls.json
+    dataset_dir: dataset/ppvehicle
+    data_fields: ['image', 'gt_bbox', 'gt_class', 'is_crowd']
+
+EvalDataset:
+  !COCODataSet
+    image_dir: ""
+    anno_path: annotations/val_all_9cls.json
+    dataset_dir: dataset/ppvehicle
+
+TestDataset:
+  !ImageFolder
+    anno_path: annotations/val_all_9cls.json
+    dataset_dir: dataset/ppvehicle
+
+TrainReader:
+  batch_size: 8
+
+epoch: 36
+LearningRate:
+  base_lr: 0.001
+  schedulers:
+    - !CosineDecay
+      max_epochs: 43
+    - !LinearWarmup
+      start_factor: 0.
+      epochs: 1
+
+PPYOLOEHead:
+  static_assigner_epoch: -1
+  nms:
+    name: MultiClassNMS
+    nms_top_k: 1000
+    keep_top_k: 100
+    score_threshold: 0.01
+    nms_threshold: 0.6

configs/ppvehicle/mot_ppyoloe_s_36e_ppvehicle.yml

+_BASE_: [
+  '../datasets/coco_detection.yml',
+  '../runtime.yml',
+  '../ppyoloe/_base_/optimizer_300e.yml',
+  '../ppyoloe/_base_/ppyoloe_crn.yml',
+  '../ppyoloe/_base_/ppyoloe_reader.yml',
+]
+log_iter: 100
+snapshot_epoch: 4
+weights: output/mot_ppyoloe_s_36e_ppvehicle/model_final
+
+pretrain_weights: https://paddledet.bj.bcebos.com/models/ppyoloe_crn_s_300e_coco.pdparams
+depth_mult: 0.33
+width_mult: 0.50
+
+num_classes: 1
+
+TrainDataset:
+  !COCODataSet
+    image_dir: ""
+    anno_path: annotations/train_all.json
+    dataset_dir: dataset/ppvehicle
+    data_fields: ['image', 'gt_bbox', 'gt_class', 'is_crowd']
+    allow_empty: true
+
+EvalDataset:
+  !COCODataSet
+    image_dir: ""
+    anno_path: annotations/val_all.json
+    dataset_dir: dataset/ppvehicle
+
+TestDataset:
+  !ImageFolder
+    anno_path: annotations/val_all.json
+    dataset_dir: dataset/ppvehicle
+
+TrainReader:
+  batch_size: 8
+
+epoch: 36
+LearningRate:
+  base_lr: 0.001
+  schedulers:
+    - !CosineDecay
+      max_epochs: 43
+    - !LinearWarmup
+      start_factor: 0.
+      epochs: 1
+
+PPYOLOEHead:
+  static_assigner_epoch: -1
+  nms:
+    name: MultiClassNMS
+    nms_top_k: 1000
+    keep_top_k: 100
+    score_threshold: 0.01
+    nms_threshold: 0.6

configs/ppvehicle/mot_ppyoloe_s_36e_ppvehicle9cls.yml

+_BASE_: [
+  '../datasets/coco_detection.yml',
+  '../runtime.yml',
+  '../ppyoloe/_base_/optimizer_300e.yml',
+  '../ppyoloe/_base_/ppyoloe_crn.yml',
+  '../ppyoloe/_base_/ppyoloe_reader.yml',
+]
+log_iter: 100
+snapshot_epoch: 4
+weights: output/mot_ppyoloe_s_36e_ppvehicle9cls/model_final
+
+pretrain_weights: https://paddledet.bj.bcebos.com/models/ppyoloe_crn_s_300e_coco.pdparams
+depth_mult: 0.33
+width_mult: 0.50
+
+num_classes: 9
+
+TrainDataset:
+  !COCODataSet
+    image_dir: ""
+    anno_path: annotations/train_all_9cls.json
+    dataset_dir: dataset/ppvehicle
+    data_fields: ['image', 'gt_bbox', 'gt_class', 'is_crowd']
+
+EvalDataset:
+  !COCODataSet
+    image_dir: ""
+    anno_path: annotations/val_all_9cls.json
+    dataset_dir: dataset/ppvehicle
+
+TestDataset:
+  !ImageFolder
+    anno_path: annotations/val_all_9cls.json
+    dataset_dir: dataset/ppvehicle
+
+TrainReader:
+  batch_size: 8
+
+epoch: 36
+LearningRate:
+  base_lr: 0.001
+  schedulers:
+    - !CosineDecay
+      max_epochs: 43
+    - !LinearWarmup
+      start_factor: 0.
+      epochs: 1
+
+PPYOLOEHead:
+  static_assigner_epoch: -1
+  nms:
+    name: MultiClassNMS
+    nms_top_k: 1000
+    keep_top_k: 100
+    score_threshold: 0.01
+    nms_threshold: 0.6

configs/ppvehicle/ppyoloe_crn_l_36e_bdd100kdet.yml

+_BASE_: [
+  '../datasets/coco_detection.yml',
+  '../runtime.yml',
+  '../ppyoloe/_base_/optimizer_300e.yml',
+  '../ppyoloe/_base_/ppyoloe_crn.yml',
+  '../ppyoloe/_base_/ppyoloe_reader.yml',
+]
+log_iter: 100
+snapshot_epoch: 4
+weights: output/ppyoloe_crn_l_36e_bdd100kdet/model_final
+
+pretrain_weights: https://paddledet.bj.bcebos.com/models/ppyoloe_crn_l_300e_coco.pdparams
+depth_mult: 1.0
+width_mult: 1.0
+
+num_classes: 10
+
+TrainDataset:
+  !COCODataSet
+    image_dir: images/100k/train
+    anno_path: labels/det_20/det_train_cocofmt.json
+    dataset_dir: dataset/bdd100k
+    data_fields: ['image', 'gt_bbox', 'gt_class', 'is_crowd']
+
+EvalDataset:
+  !COCODataSet
+    image_dir: images/100k/val
+    anno_path: labels/det_20/det_val_cocofmt.json
+    dataset_dir: dataset/bdd100k
+
+TestDataset:
+  !ImageFolder
+    anno_path: labels/det_20/det_val_cocofmt.json
+    dataset_dir: dataset/bdd100k
+
+TrainReader:
+  batch_size: 8
+
+epoch: 36
+LearningRate:
+  base_lr: 0.001
+  schedulers:
+    - !CosineDecay
+      max_epochs: 43
+    - !LinearWarmup
+      start_factor: 0.
+      epochs: 1
+
+PPYOLOEHead:
+  static_assigner_epoch: -1
+  nms:
+    name: MultiClassNMS
+    nms_top_k: 1000
+    keep_top_k: 100
+    score_threshold: 0.01
+    nms_threshold: 0.6

configs/ppvehicle/ppyoloe_crn_l_36e_bdd100kmot.yml

+_BASE_: [
+  '../datasets/coco_detection.yml',
+  '../runtime.yml',
+  '../ppyoloe/_base_/optimizer_300e.yml',
+  '../ppyoloe/_base_/ppyoloe_crn.yml',
+  '../ppyoloe/_base_/ppyoloe_reader.yml',
+]
+log_iter: 100
+snapshot_epoch: 4
+weights: output/ppyoloe_crn_l_36e_bdd100kmot/model_final
+
+pretrain_weights: https://paddledet.bj.bcebos.com/models/ppyoloe_crn_l_300e_coco.pdparams
+depth_mult: 1.0
+width_mult: 1.0
+
+num_classes: 8
+
+TrainDataset:
+  !COCODataSet
+    image_dir: ""
+    anno_path: annotations/train.json
+    dataset_dir: dataset/bdd100k
+    data_fields: ['image', 'gt_bbox', 'gt_class', 'is_crowd']
+
+EvalDataset:
+  !COCODataSet
+    image_dir: ""
+    anno_path: annotations/val.json
+    dataset_dir: dataset/bdd100k
+
+TestDataset:
+  !ImageFolder
+    anno_path: annotations/val.json
+    dataset_dir: dataset/bdd100k
+
+TrainReader:
+  batch_size: 8
+
+epoch: 36
+LearningRate:
+  base_lr: 0.001
+  schedulers:
+    - !CosineDecay
+      max_epochs: 43
+    - !LinearWarmup
+      start_factor: 0.
+      epochs: 1
+
+PPYOLOEHead:
+  static_assigner_epoch: -1
+  nms:
+    name: MultiClassNMS
+    nms_top_k: 1000
+    keep_top_k: 100
+    score_threshold: 0.01
+    nms_threshold: 0.6

configs/ppvehicle/ppyoloe_crn_l_36e_uadetrac.yml

+_BASE_: [
+  '../datasets/coco_detection.yml',
+  '../runtime.yml',
+  '../ppyoloe/_base_/optimizer_300e.yml',
+  '../ppyoloe/_base_/ppyoloe_crn.yml',
+  '../ppyoloe/_base_/ppyoloe_reader.yml',
+]
+log_iter: 100
+snapshot_epoch: 4
+weights: output/ppyoloe_crn_l_36e_uadetrac/model_final
+
+pretrain_weights: https://paddledet.bj.bcebos.com/models/ppyoloe_crn_l_300e_coco.pdparams
+depth_mult: 1.0
+width_mult: 1.0
+
+num_classes: 4
+
+TrainDataset:
+  !COCODataSet
+    image_dir: train
+    anno_path: annotations/train.json
+    dataset_dir: dataset/uadetrac
+    data_fields: ['image', 'gt_bbox', 'gt_class', 'is_crowd']
+
+EvalDataset:
+  !COCODataSet
+    image_dir: val
+    anno_path: annotations/test.json
+    dataset_dir: dataset/uadetrac
+
+TestDataset:
+  !ImageFolder
+    anno_path: annotations/test.json
+    dataset_dir: dataset/uadetrac
+
+TrainReader:
+  batch_size: 8
+
+epoch: 36
+LearningRate:
+  base_lr: 0.001
+  schedulers:
+    - !CosineDecay
+      max_epochs: 43
+    - !LinearWarmup
+      start_factor: 0.
+      epochs: 1
+
+PPYOLOEHead:
+  static_assigner_epoch: -1
+  nms:
+    name: MultiClassNMS
+    nms_top_k: 1000
+    keep_top_k: 100
+    score_threshold: 0.01
+    nms_threshold: 0.6

configs/ppyoloe/README.md

@@ -177,6 +177,17 @@ Model | AP | AP<sub>50</sub>
 - *Due to the limited time, more verification results will be supplemented in the future. You are also welcome to contribute to PP-YOLOE*
 
 
+### Feature Models
+
+The PaddleDetection team provides configs and weights of various feature detection models based on PP-YOLOE, which users can download for use:
+
+|Scenarios | Related Datasets | Links|
+| :--------: | :---------: | :------: |
+|Pedestrian Detection | CrowdHuman | [pphuman](../pphuman) |
+|Vehicle Detection | BDD100K，UA-DETRAC | [ppvehicle](../ppvehicle) |
+|Small Object Detection | VisDrone | [visdrone](../visdrone) |
+
+
 ## Appendix
 
 Ablation experiments of PP-YOLOE.

configs/ppyoloe/README_cn.md

@@ -178,6 +178,18 @@ CUDA_VISIBLE_DEVICES=0 python deploy/python/infer.py --model_dir=output_inferenc
 - 以上模型训练均采用官方提供的默认参数，并且加载COCO预训练参数
 - *由于人力/时间有限，后续将会持续补充更多验证结果，也欢迎各位开源用户贡献，共同优化PP-YOLOE*
 
+
+### 垂类模型应用
+
+PaddleDetection团队提供了基于PP-YOLOE的各种垂类检测模型的配置文件和权重，用户可以下载进行使用：
+
+|     场景    |    相关数据集    |    链接   |
+| :--------: | :---------: | :------: |
+|  行人检测   | CrowdHuman   |   [pphuman](../pphuman)  |
+|  车辆检测   | BDD100K，UA-DETRAC  |  [ppvehicle](../ppvehicle)   |
+|  小目标检测 | VisDrone     |  [visdrone](../visdrone)   |
+
+
 ## 附录
 
 PP-YOLOE消融实验