- We use a multi-scale evaluation strategy to get the mAP in `Easy/Medium/Hard Set`. Please refer to the [evaluation on the WIDER FACE dataset](#Evaluated-on-the-WIDER-FACE-Dataset) for details.
- We use a multi-scale evaluation strategy to get the mAP in `Easy/Medium/Hard Set`. Please refer to the [evaluation on the WIDER FACE dataset](#Evaluated-on-the-WIDER-FACE-Dataset) for details.
@@ -17,7 +17,7 @@ The network for detecting vehicles is YOLOv3, the backbone of which is Dacknet53
...
@@ -17,7 +17,7 @@ The network for detecting vehicles is YOLOv3, the backbone of which is Dacknet53
### 2. Configuration for training
### 2. Configuration for training
PaddleDetection provides users with a configuration file [yolov3_darknet53_270e_coco.yml](https://github.com/PaddlePaddle/PaddleDetection/blob/develop/configs/yolov3/yolov3_darknet53_270e_coco.yml) to train YOLOv3 on the COCO dataset, compared with this file, we modify some parameters as followed to conduct the training for pedestrian detection:
PaddleDetection provides users with a configuration file [yolov3_darknet53_270e_coco.yml](https://github.com/PaddlePaddle/PaddleDetection/blob/release/2.3/configs/yolov3/yolov3_darknet53_270e_coco.yml) to train YOLOv3 on the COCO dataset, compared with this file, we modify some parameters as followed to conduct the training for pedestrian detection:
- PP-YOLO is trained on COCO train2017 dataset and evaluated on val2017 & test-dev2017 dataset，Box AP<sup>test</sup> is evaluation results of `mAP(IoU=0.5:0.95)`.
- PP-YOLO is trained on COCO train2017 dataset and evaluated on val2017 & test-dev2017 dataset，Box AP<sup>test</sup> is evaluation results of `mAP(IoU=0.5:0.95)`.
- PP-YOLO used 8 GPUs for training and mini-batch size as 24 on each GPU, if GPU number and mini-batch size is changed, learning rate and iteration times should be adjusted according [FAQ](https://github.com/PaddlePaddle/PaddleDetection/blob/develop/docs/tutorials/FAQ.md).
- PP-YOLO used 8 GPUs for training and mini-batch size as 24 on each GPU, if GPU number and mini-batch size is changed, learning rate and iteration times should be adjusted according [FAQ](https://github.com/PaddlePaddle/PaddleDetection/blob/release/2.3/docs/tutorials/FAQ.md).
- PP-YOLO inference speed is tesed on single Tesla V100 with batch size as 1, CUDA 10.2, CUDNN 7.5.1, TensorRT 5.1.2.2 in TensorRT mode.
- PP-YOLO inference speed is tesed on single Tesla V100 with batch size as 1, CUDA 10.2, CUDNN 7.5.1, TensorRT 5.1.2.2 in TensorRT mode.
- PP-YOLO FP32 inference speed testing uses inference model exported by `tools/export_model.py` and benchmarked by running `depoly/python/infer.py` with `--run_benchmark`. All testing results do not contains the time cost of data reading and post-processing(NMS), which is same as [YOLOv4(AlexyAB)](https://github.com/AlexeyAB/darknet) in testing method.
- PP-YOLO FP32 inference speed testing uses inference model exported by `tools/export_model.py` and benchmarked by running `depoly/python/infer.py` with `--run_benchmark`. All testing results do not contains the time cost of data reading and post-processing(NMS), which is same as [YOLOv4(AlexyAB)](https://github.com/AlexeyAB/darknet) in testing method.
- TensorRT FP16 inference speed testing exclude the time cost of bounding-box decoding(`yolo_box`) part comparing with FP32 testing above, which means that data reading, bounding-box decoding and post-processing(NMS) is excluded(test method same as [YOLOv4(AlexyAB)](https://github.com/AlexeyAB/darknet) too)
- TensorRT FP16 inference speed testing exclude the time cost of bounding-box decoding(`yolo_box`) part comparing with FP32 testing above, which means that data reading, bounding-box decoding and post-processing(NMS) is excluded(test method same as [YOLOv4(AlexyAB)](https://github.com/AlexeyAB/darknet) too)
...
@@ -69,21 +69,21 @@ PP-YOLO and PP-YOLOv2 improved performance and speed of YOLOv3 with following me
...
@@ -69,21 +69,21 @@ PP-YOLO and PP-YOLOv2 improved performance and speed of YOLOv3 with following me
| Model | GPU number | images/GPU | Model Size | input shape | Box AP<sup>val</sup> | Box AP50<sup>val</sup> | Kirin 990 1xCore(FPS) | download | config |
| Model | GPU number | images/GPU | Model Size | input shape | Box AP<sup>val</sup> | Box AP50<sup>val</sup> | Kirin 990 1xCore(FPS) | download | config |
- PP-YOLO_MobileNetV3 is trained on COCO train2017 datast and evaluated on val2017 dataset，Box AP<sup>val</sup> is evaluation results of `mAP(IoU=0.5:0.95)`, Box AP<sup>val</sup> is evaluation results of `mAP(IoU=0.5)`.
- PP-YOLO_MobileNetV3 is trained on COCO train2017 datast and evaluated on val2017 dataset，Box AP<sup>val</sup> is evaluation results of `mAP(IoU=0.5:0.95)`, Box AP<sup>val</sup> is evaluation results of `mAP(IoU=0.5)`.
- PP-YOLO_MobileNetV3 used 4 GPUs for training and mini-batch size as 32 on each GPU, if GPU number and mini-batch size is changed, learning rate and iteration times should be adjusted according [FAQ](https://github.com/PaddlePaddle/PaddleDetection/blob/develop/docs/tutorials/FAQ.md).
- PP-YOLO_MobileNetV3 used 4 GPUs for training and mini-batch size as 32 on each GPU, if GPU number and mini-batch size is changed, learning rate and iteration times should be adjusted according [FAQ](https://github.com/PaddlePaddle/PaddleDetection/blob/release/2.3/docs/tutorials/FAQ.md).
- PP-YOLO_MobileNetV3 inference speed is tested on Kirin 990 with 1 thread.
- PP-YOLO_MobileNetV3 inference speed is tested on Kirin 990 with 1 thread.
### PP-YOLO tiny
### PP-YOLO tiny
| Model | GPU number | images/GPU | Model Size | Post Quant Model Size | input shape | Box AP<sup>val</sup> | Kirin 990 4xCore(FPS) | download | config | post quant model |
| Model | GPU number | images/GPU | Model Size | Post Quant Model Size | input shape | Box AP<sup>val</sup> | Kirin 990 4xCore(FPS) | download | config | post quant model |
@@ -212,7 +212,7 @@ Optimizing method and ablation experiments of PP-YOLO compared with YOLOv3.
...
@@ -212,7 +212,7 @@ Optimizing method and ablation experiments of PP-YOLO compared with YOLOv3.
- Performance and inference spedd are measure with input shape as 608
- Performance and inference spedd are measure with input shape as 608
- All models are trained on COCO train2017 datast and evaluated on val2017 & test-dev2017 dataset，`Box AP` is evaluation results as `mAP(IoU=0.5:0.95)`.
- All models are trained on COCO train2017 datast and evaluated on val2017 & test-dev2017 dataset，`Box AP` is evaluation results as `mAP(IoU=0.5:0.95)`.
- Inference speed is tested on single Tesla V100 with batch size as 1 following test method and environment configuration in benchmark above.
- Inference speed is tested on single Tesla V100 with batch size as 1 following test method and environment configuration in benchmark above.
-[YOLOv3-DarkNet53](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/yolov3/yolov3_darknet53_270e_coco.yml) with mAP as 39.0 is optimized YOLOv3 model in PaddleDetection，see [YOLOv3](https://github.com/PaddlePaddle/PaddleDetection/blob/develop/configs/yolov3/README.md) for details.
-[YOLOv3-DarkNet53](https://github.com/PaddlePaddle/PaddleDetection/tree/release/2.3/configs/yolov3/yolov3_darknet53_270e_coco.yml) with mAP as 39.0 is optimized YOLOv3 model in PaddleDetection，see [YOLOv3](https://github.com/PaddlePaddle/PaddleDetection/blob/develop/configs/yolov3/README.md) for details.
| Backbone | Network type | Number of images per GPU | Learning rate strategy | Inferring time(fps) | Box AP | Mask AP | Download | Configuration File |
| Backbone | Network type | Number of images per GPU | Learning rate strategy | Inferring time(fps) | Box AP | Mask AP | Download | Configuration File |
- The above V100 prediction delay non-quantified model is tested by TensorRT FP32, and the quantified model is tested by TensorRT INT8, and both of them include NMS time.
- The above V100 prediction delay non-quantified model is tested by TensorRT FP32, and the quantified model is tested by TensorRT INT8, and both of them include NMS time.
...
@@ -151,8 +151,8 @@ Description:
...
@@ -151,8 +151,8 @@ Description:
| Model | Compression Strategy | Input Size | Box AP | Download | Model Configuration File | Compression Strategy Configuration File |
| Model | Compression Strategy | Input Size | Box AP | Download | Model Configuration File | Compression Strategy Configuration File |
COCO数据集作为目标检测任务的训练目标难度更大，意味着teacher网络会预测出更多的背景bbox，如果直接用teacher的预测输出作为student学习的`soft label`会有严重的类别不均衡问题。解决这个问题需要引入新的方法，详细背景请参考论文:[Object detection at 200 Frames Per Second](https://arxiv.org/abs/1805.06361)。
COCO数据集作为目标检测任务的训练目标难度更大，意味着teacher网络会预测出更多的背景bbox，如果直接用teacher的预测输出作为student学习的`soft label`会有严重的类别不均衡问题。解决这个问题需要引入新的方法，详细背景请参考论文:[Object detection at 200 Frames Per Second](https://arxiv.org/abs/1805.06361)。
**Attention:** Due to the overall upgrade of the dynamic graph framework, the weighting model published by PaddleDetection of PAF Net needs to be evaluated with a --bias field, for example
**Attention:** Due to the overall upgrade of the dynamic graph framework, the weighting model published by PaddleDetection of PAF Net needs to be evaluated with a --bias field, for example
@@ -17,7 +17,7 @@ The network for detecting vehicles is YOLOv3, the backbone of which is Dacknet53
...
@@ -17,7 +17,7 @@ The network for detecting vehicles is YOLOv3, the backbone of which is Dacknet53
### 2. Configuration for training
### 2. Configuration for training
PaddleDetection provides users with a configuration file [yolov3_darknet53_270e_coco.yml](https://github.com/PaddlePaddle/PaddleDetection/blob/develop/configs/yolov3/yolov3_darknet53_270e_coco.yml) to train YOLOv3 on the COCO dataset, compared with this file, we modify some parameters as followed to conduct the training for vehicle detection:
PaddleDetection provides users with a configuration file [yolov3_darknet53_270e_coco.yml](https://github.com/PaddlePaddle/PaddleDetection/blob/release/2.3/configs/yolov3/yolov3_darknet53_270e_coco.yml) to train YOLOv3 on the COCO dataset, compared with this file, we modify some parameters as followed to conduct the training for vehicle detection:
