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

NotFoundError: driver_100_30frame/05251517_0433.MP4/00000.jpg #62

Closed YilmazOrkun closed 5 years ago

YilmazOrkun commented 5 years ago

Hello, unfortunately I have problems with image paths. I already did read closed issues about that and the suggestion was to use absolute path. But where do I enter the absolute path? In the test.txt list file?

Thanks in advance!

cardwing commented 5 years ago

@YilmazOrkun, just print the path and check if it exists.

YilmazOrkun commented 5 years ago

Dear @cardwing, unfortunately I don't understand your answer. Could you please more specific.

My test.txt looks like this:

~/SCNN/data/CULane/driver_100_30frame/05251517_0433.MP4/00000.jpg ~/SCNN/data/CULane/driver_100_30frame/05251517_0433.MP4/00030.jpg ~/SCNN/data/CULane/driver_100_30frame/05251517_0433.MP4/00060.jpg

So I did use absolute path. And I call testing like this:

CUDA_VISIBLE_DEVICES="0" python tools/test_lanenet.py --weights_path ~/SCNN-Tensorflow/lane-detection-model/culane_lanenet_vgg_2018-12-01-14-38-37.ckpt-10000 --image_path ~/SCNN/data/CULane/list/test.txt --save_dir ~/SCNN-Tensorflow/lane-detection-model/predictions

But the images still can't be found and I get the error: NotFoundError (see above for traceback): home/mrk/SCNN/data/CULane/driver_100_30frame/05251517_0433.MP4/00000.jpg

Thanks in advance

cardwing commented 5 years ago

Can you cd home/mrk/SCNN/data/CULane/driver_100_30frame/05251517_0433.MP4? It should be '/home/...' instead of 'home/...'.