The lane evaluation in CULane is accelerated about 20 x

cardwing / Codes-for-Lane-Detection

Learning Lightweight Lane Detection CNNs by Self Attention Distillation (ICCV 2019)

MIT License

1.04k stars 333 forks source link

The lane evaluation in CULane is accelerated about 20 x #163

Closed cardwing closed 5 years ago

cardwing commented 5 years ago

The newly added code for the lane evaluation in CULane can fasten the evaluation process about 20 x. Just compile with make in ERFNet-CULane-PyTorch/tools/lane_evaluation/ (make clean first). If you use OpenCV 2.4.8, please remove opencv_imgcodecs in ERFNet-CULane-PyTorch/tools/lane_evaluation/Makefile and change #include <opencv2/imgproc.hpp> to #include <opencv2/imgproc/imgproc.hpp> in ERFNet-CULane-PyTorch/tools/lane_evaluation/include/lane_compare.hpp. Thanks @atranitell

atranitell commented 5 years ago

Yeah, the newest lane evaluation tool could directly used in OpenCV 3 above. Thank @cardwing points that.