littlemountainman / modeld Goto Github PK
View Code? Open in Web Editor NEWSelf driving car lane and path detection
Home Page: https://www.youtube.com/watch?v=UFQQbTYH9hI
Self driving car lane and path detection
Home Page: https://www.youtube.com/watch?v=UFQQbTYH9hI
My docker file:
FROM ubuntu:20.04
RUN apt-get -y update
RUN apt-get -y install git vim curl python3-pip
RUN git clone https://github.com/littlemountainman/modeld.git
WORKDIR /modeld
COPY sample.hevc /modeld/
RUN pip3 install -r requirements.txt
CMD ["python3", "main.py sample.hevc"]
gives me error:
#9 8.378 Downloading botocore-1.15.26-py2.py3-none-any.whl (6.0 MB)
#9 8.821 Collecting bs4==0.0.1
#9 8.840 Downloading bs4-0.0.1.tar.gz (1.1 kB)
#9 9.030 Collecting cachetools==4.0.0
#9 9.047 Downloading cachetools-4.0.0-py3-none-any.whl (10 kB)
#9 9.089 ERROR: Could not find a version that satisfies the requirement carla==0.9.7 (from -r requirements.txt (line 17)) (from versions: 0.9.5)
#9 9.247 ERROR: No matching distribution found for carla==0.9.7 (from -r requirements.txt (line 17))
------
same error when I try locally on my MacBook Pro in a virtual environment. Appreciate help!
I have a question that why supercombo.keras output points in space? I think the lane line label is in image coordinate, is it?
Hi this is a great implementation. I tried it, really cool. But compared to the current model (new torch model) outputs, the keras model has a relatively large variance, especially for the speed and acceleration.
I'm wondering do you have plans to implement the new torch model?
I have watched your youtube and will try it out on my PC.
what is your discord channel?
My interest is to explore whether this can be ported to rk3399pro which has NPU accelerator and run in realtime.
Thanks,
I converted an mp4 video to hevc format using
ffmpeg -i Driving.mp4 -c:v libx265 -vtag hvc1 output.hevc
as well as
ffmpeg -i Driving.mp4 -c:v libx265 -vtag hvc1 output.mp4
But for both the files I get Data Unreadable error when running the minimal.py file. Also the sample data file mentioned in the readme is not available for download. Help please.
Thank you.
hi, Author.
carla==0.9.7 how to get?
When I carry out orders( pip3 install -r requirements.txt )
ERROR: Could not find a version that satisfies the requirement carla==0.9.7 (from -r requirements.txt (line 17)) (from versions: 0.9.3, 0.9.4, 0.9.5)
ERROR: No matching distribution found for carla==0.9.7 (from -r requirements.txt (line 17))
https://pypi.org/search/?q=carla is carla 0.9.5
I get this error whe running main.py,
I am on windows with python 3.7
Traceback (most recent call last):
File "main.py", line 24, in
fr = FrameReader(camerafile)
File "C:\Users\asus\Desktop\Self Driving\comma supercombo\modeld-master\tools\lib\framereader.py", line 679, in FrameReader
index_data = get_video_index(fn, frame_type, cache_prefix)
File "C:\Users\asus\Desktop\Self Driving\comma supercombo\modeld-master\tools\lib\framereader.py", line 215, in get_video_index
return pickle.load(cache_file)
_pickle.UnpicklingError: invalid load key, '\x00'.
Will it be possible to overlay lane/driving prediction to the original video instead of displaying on a separate window?
There is a repository, superdrive below which is similar to your code and I believe it has better memory management. You may want to check it out
https://github.com/NamoDev/SuperDrive
Thanks,
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ 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. ๐๐๐
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 โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.