YOLOP
YOLOP copied to clipboard
hustvl
You Only Look Once for Panopitic Driving Perception.（MIR2022）
I have a requirement for Tensorrt implementation of this lane detection in my Jetson Nano/ Jetson Orin nano device. I can see that there is Tensrort C++ API for it...
Description I was convert wts to trt, model building successfully,but still have error. Please give me help,thanks! Environment System: AGX Orin TensorRT Version: 8.5.2.2 CUDA Version: 11.4 CUDNN Version: 8.6...
the model tain speed is very slow, it will cost about 20 minutes on the whole BDD100K datasets with 2 GPUS, I tried modified the `num_workers` and `pin_memory` is set...
训练配置如下： (torch171) lpj@252-2titanx:~/csn_work/YOLOP$ python -m torch.distributed.launch --nproc_per_node=2 tools/train.py ***************************************** Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune...
如题，采用多卡训练，如果冻结部分网络，程序会卡死。 比如，我在冻结seg分支后，训练det分支，会在设置anchors这块卡死。 有没有遇到上述问题，该如何解决
use release weight to eval, Lane Detection Result is wrong. 
Hi, I am reproducing your work, but I have encountered a problem. It seems that you used different values for the --conf_thres and --iou_thres hyperparameters in demo.py and test.py. After...
