- FairMOT HRNetV2-W18 used 8 GPUs for training and mini-batch size as 4 on each GPU, and trained for 30 epoches. Only ImageNet pre-train model is used, and the optimizer adopts Momentum. The crowdhuman dataset is added to the train-set during training.
- FairMOT HRNetV2-W18 used 8 GPUs for training and mini-batch size as 4 on each GPU, and trained for 30 epochs. Only ImageNet pre-train model is used, and the optimizer adopts Momentum. The crowdhuman dataset is added to the train-set during training.
- MOTA is the average MOTA of 10 catecories in the VisDrone2019 MOT dataset, and its value is also equal to the average MOTA of all the evaluated video sequences. Here we provide the download [link](https://bj.bcebos.com/v1/paddledet/data/mot/visdrone_mcmot.zip) of the dataset.
- MOTA is the average MOTA of 10 categories in the VisDrone2019 MOT dataset, and its value is also equal to the average MOTA of all the evaluated video sequences. Here we provide the download [link](https://bj.bcebos.com/v1/paddledet/data/mot/visdrone_mcmot.zip) of the dataset.
- MCFairMOT used 4 GPUs for training 30 epoches. The batch size is 6 on each GPU for MCFairMOT DLA-34, and 8 for MCFairMOT HRNetV2-W18.
- MCFairMOT used 4 GPUs for training 30 epochs. The batch size is 6 on each GPU for MCFairMOT DLA-34, and 8 for MCFairMOT HRNetV2-W18.
- MOTA is the average MOTA of 4 catecories in the VisDrone Vehicle dataset, and this dataset is extracted from the VisDrone2019 MOT dataset, here we provide the download [link](https://bj.bcebos.com/v1/paddledet/data/mot/visdrone_mcmot_vehicle.zip).
- MOTA is the average MOTA of 4 categories in the VisDrone Vehicle dataset, and this dataset is extracted from the VisDrone2019 MOT dataset, here we provide the download [link](https://bj.bcebos.com/v1/paddledet/data/mot/visdrone_mcmot_vehicle.zip).
- The tracker used in MCFairMOT model here is ByteTracker.
- The tracker used in MCFairMOT model here is ByteTracker.
### MCFairMOT off-line quantization results on VisDrone Vehicle val-set
### MCFairMOT off-line quantization results on VisDrone Vehicle val-set
@@ -60,7 +60,7 @@ PP-YOLO and PP-YOLOv2 improved performance and speed of YOLOv3 with following me
...
@@ -60,7 +60,7 @@ PP-YOLO and PP-YOLOv2 improved performance and speed of YOLOv3 with following me
- 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).
- 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).
- 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 tested 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)
- If you set `--run_benchmark=True`，you should install these dependencies at first, `pip install pynvml psutil GPUtil`.
- If you set `--run_benchmark=True`，you should install these dependencies at first, `pip install pynvml psutil GPUtil`.
For inference deployment or benchmard, model exported with `tools/export_model.py` should be used and perform inference with Paddle inference library with following commands:
For inference deployment or benchmard, model exported with `tools/export_model.py` should be used and perform inference with Paddle inference library with following commands:
- PaddleDetection supports multi-machine distribued training, you can refer to [DistributedTraining tutorial](../../docs/tutorials/DistributedTraining_en.md).
- PaddleDetection supports multi-machine distributed training, you can refer to [DistributedTraining tutorial](../../docs/tutorials/DistributedTraining_en.md).
- Here, we use [VisDrone](https://github.com/VisDrone/VisDrone-Dataset) dataset, and to detect 9 objects including `person, bicycles, car, van, truck, tricyle, awning-tricyle, bus, motor`.
- Here, we use [VisDrone](https://github.com/VisDrone/VisDrone-Dataset) dataset, and to detect 9 objects including `person, bicycles, car, van, truck, tricycle, awning-tricycle, bus, motor`.
- Above models trained using official default config, and load pretrained parameters on COCO dataset.
- Above models trained using official default config, and load pretrained parameters on COCO dataset.
-*Due to the limited time, more verification results will be supplemented in the future. You are also welcome to contribute to PP-YOLOE*
-*Due to the limited time, more verification results will be supplemented in the future. You are also welcome to contribute to PP-YOLOE*
- The ResNet50-FPN are trained on COCO train2017 with 8 GPUs. Both ResNet101-FPN and ResNet50-FPN with [FGD](../slim/distill/README.md) are trained on COCO train2017 with 4 GPUs.
- The ResNet50-FPN are trained on COCO train2017 with 8 GPUs. Both ResNet101-FPN and ResNet50-FPN with [FGD](../slim/distill/README.md) are trained on COCO train2017 with 4 GPUs.
- All above models are evaludated on val2017. Box AP=`mAP(IoU=0.5:0.95)`.
- All above models are evaluated on val2017. Box AP=`mAP(IoU=0.5:0.95)`.
