ericlavigne / carnd-detect-lane-lines-and-vehicles Goto Github PK
View Code? Open in Web Editor NEWUse segmentation networks to recognize lane lines and vehicles. Infer position and curvature of lane lines relative to self.
Use segmentation networks to recognize lane lines and vehicles. Infer position and curvature of lane lines relative to self.
hello Eric,
how to convert the images to pixen format, Can you please share the script?
I have implemented SCNN using Tensorflow and put the full codes here. You can test the code in popular lane detection benchmarks like TuSimple, CULane and BDD100K or your custom dataset with minor modification. Welcome to raising issues if you have problems in reproducing the results. My code is based on LaneNet and SCNN-Torch.
Hi,
Thank you for the sharing. May I know how I can set the batch size since it will run out of memory if more images are trained. Thank you.
File "main.py", line 532, in
main()
File "main.py", line 526, in main
transform_image_files(lambda img: video_processor(lane_model=lane_model,car_model=car_model,calibration=calibration).process_image(img),
File "main.py", line 57, in transform_image_files
dst_img = transformation(img)
File "main.py", line 526, in
transform_image_files(lambda img: video_processor(lane_model=lane_model,car_model=car_model,calibration=calibration).process_image(img),
File "main.py", line 451, in process_image
markings = image_to_prediction(undistorted, self.lane_model, lane_settings)
File "main.py", line 234, in image_to_prediction
result = uncrop_scale(result,opt)
File "main.py", line 140, in uncrop_scale
img = uncrop(img,opt)
File "main.py", line 115, in uncrop
frame[opt['crop_min_y']:opt['crop_max_y'], opt['crop_min_x']:opt['crop_max_x'], 0:3] = img
ValueError: could not broadcast input array from shape (2,2,3) into shape (246,880,3)
Can anyone help with this issue? @ericlavigne @Brok-Bucholtz @ryan-keenan
hello,i am using the program to process my own videos which are already resize to 1280*720,and error below is appearing.
`
File "C:\Users\10401\AppData\Local\conda\conda\envs\TensorFlow\lib\site-packages\numpy\core_methods.py", line 29, in _amin
return umr_minimum(a, axis, None, out, keepdims)
ValueError: zero-size array to reduction operation minimum which has no identity`
i am not familiar with python.So why it appears and how can i solve the problem?
Hello,
In the step of running the project, I get the following error.
File "main.py", line 407, in annotate_original_image
if lane_markings_img != None:
The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()
Traceback (most recent call last):
File "main.py", line 529, in
main()
File "main.py", line 522, in main
'output_images/birds_eye_lines')
File "main.py", line 54, in transform_image_files
dst_img = transformation(img)
File "main.py", line 380, in convert_lane_heatmap_to_lane_lines_image
lines = fit_parabolas_to_lane_centroids(centroids)
File "main.py", line 332, in fit_parabolas_to_lane_centroids
min_y = np.amin(y_vals)
File "/home/syp/CarND-Detect-Lane-Lines-And-Vehicles/env/lib/python3.6/site-packages/numpy/core/fromnumeric.py", line 2352, in amin
out=out, **kwargs)
File "/home/syp/CarND-Detect-Lane-Lines-And-Vehicles/env/lib/python3.6/site-packages/numpy/core/_methods.py", line 29, in _amin
return umr_minimum(a, axis, None, out, keepdims)
ValueError: zero-size array to reduction operation minimum which has no identity
I change the test_images's image to my image,and the size was 1280*720, but it has a error.
While you mention below, I was trying to make one for my own data. However, I couldn't find a way like your precise marked on lane and car. Would you please show some tips for it?
"I copied each of these images to the training directory, for annotation. I converted the images to Pixen format, added layers to represent lane markings and cars, and created image masks in those layers to indicate the locations of lane markings and cars. I saved each layer separately with filenames ending in "x", "lanes", and "cars" so they could easily be imported into Python for training convolutional neural networks."
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.