Szal, egy nyúl ma azt hitte, hogy felzabálja a büszke bokrunkat a kertben. Aztán a shishi odoshi helyrerakta.
2015. augusztus 13., csütörtök
2015. augusztus 6., csütörtök
A new update of my artificial intelligence for Sony ERS-7 robots
Here comes the new update of my AiBO+ AIBOWare for Sony ERS-7: AiBO+ 3.1 - Let's Play!
New changes in this release:
- New sound profiles with voice acting (text-to-speech, Marissa, Gillian)
- A game can be played with AIBO
- Reworked pick-up mode/fall over/poking detection
- Configurable adaptive volume control
- Quicker responses for petting
- Less stress on the motors
- Many small fixes
The AiBO+ AIBOWare must be installed in a PMS stick and an AiBO+ Client application can be run under Windows, Ubuntu Linux and Android.
The installation instructions and a detailed user guide are available on this page:
http://aiboplus.sourceforge.net/userguide.html
New changes in this release:
- New sound profiles with voice acting (text-to-speech, Marissa, Gillian)
- A game can be played with AIBO
- Reworked pick-up mode/fall over/poking detection
- Configurable adaptive volume control
- Quicker responses for petting
- Less stress on the motors
- Many small fixes
The AiBO+ AIBOWare must be installed in a PMS stick and an AiBO+ Client application can be run under Windows, Ubuntu Linux and Android.
The installation instructions and a detailed user guide are available on this page:
http://aiboplus.sourceforge.net/userguide.html
2015. augusztus 3., hétfő
Second astrophoto - Shapley 1
Shapley 1 (planetary nebula)
Captured with telescope T32 of iTelescope network, Side Springs Observatory, Australia.
Capture parameters: LRGB, 14x120, 4x120, 4x120, 4x120.
It is my second trial with astroimaging. I was a bit surprised that I got better results when I tried to work with uncalibrated images. I tried various filters, but none of them given reasonable results with shutter speed 300 seconds.
Captured with telescope T32 of iTelescope network, Side Springs Observatory, Australia.
Capture parameters: LRGB, 14x120, 4x120, 4x120, 4x120.
It is my second trial with astroimaging. I was a bit surprised that I got better results when I tried to work with uncalibrated images. I tried various filters, but none of them given reasonable results with shutter speed 300 seconds.
2015. május 24., vasárnap
Conversion script for FujiFilm FinePix W3 3D camera videos (avi) to SBS mp4 under Linux
When the following script is run, it looks for avi files in the current folder whose name starts with DSC... and their format is motions jpeg.
The left/right channels of the original W3 video are extracted and a final side-by-side video is created. All new videos will be in mp4 format.
Here it is the script:
#!/bin/bash
file_pattern="$@"
if [ ! $file_pattern ]; then
file_pattern="DSC*.avi DSC*.AVI"
fi
for file in $(ls -f $file_pattern 2> /dev/null)
do
echo Checking $file
if [[ "$(file $file)" == *"video: Motion JPEG"* ]]; then
echo Processing $file
filenamebase=${file%.*}
echo Create left video: "$filenamebase"_l.mp4
ffmpeg -loglevel error -i $file -f mp4 -acodec mp3 -map 0:v:1 -map 0:a:0 "$filenamebase"_l.mp4
echo Create right video: "$filenamebase"_r.mp4
ffmpeg -loglevel error -i $file -f mp4 -acodec mp3 -map 0:v:0 -map 0:a:0 "$filenamebase"_r.mp4
echo Join left/right videos to a half SBS video: "$filenamebase"_sbs.mp4
ffmpeg -loglevel error -i "$filenamebase"_l.mp4 -i "$filenamebase"_r.mp4 \
-filter_complex "pad=in_w*2:in_h, overlay=main_w/2:0, scale=in_w/2:in_h" \
-f mp4 -b:v 6000k -acodec copy -map 0:a:0 "$filenamebase"_sbs.mp4
echo Rename left video to 2d version: "$filenamebase"_l.mp4 to "$filenamebase"_2d.mp4
mv "$filenamebase"_l.mp4 "$filenamebase"_2d.mp4
fi
done
The left/right channels of the original W3 video are extracted and a final side-by-side video is created. All new videos will be in mp4 format.
Here it is the script:
#!/bin/bash
file_pattern="$@"
if [ ! $file_pattern ]; then
file_pattern="DSC*.avi DSC*.AVI"
fi
for file in $(ls -f $file_pattern 2> /dev/null)
do
echo Checking $file
if [[ "$(file $file)" == *"video: Motion JPEG"* ]]; then
echo Processing $file
filenamebase=${file%.*}
echo Create left video: "$filenamebase"_l.mp4
ffmpeg -loglevel error -i $file -f mp4 -acodec mp3 -map 0:v:1 -map 0:a:0 "$filenamebase"_l.mp4
echo Create right video: "$filenamebase"_r.mp4
ffmpeg -loglevel error -i $file -f mp4 -acodec mp3 -map 0:v:0 -map 0:a:0 "$filenamebase"_r.mp4
echo Join left/right videos to a half SBS video: "$filenamebase"_sbs.mp4
ffmpeg -loglevel error -i "$filenamebase"_l.mp4 -i "$filenamebase"_r.mp4 \
-filter_complex "pad=in_w*2:in_h, overlay=main_w/2:0, scale=in_w/2:in_h" \
-f mp4 -b:v 6000k -acodec copy -map 0:a:0 "$filenamebase"_sbs.mp4
echo Rename left video to 2d version: "$filenamebase"_l.mp4 to "$filenamebase"_2d.mp4
mv "$filenamebase"_l.mp4 "$filenamebase"_2d.mp4
fi
done
2015. április 26., vasárnap
First public release of my artificial intelligence for Sony ERS-7 robot dogs
After years of development, I created the first public version of my AiBO+ AIBOWare for Sony ERS-7:
AiBO+ 3.0 - AIBO Rebirth release.
A feature video can be watched:
The same video in 3D:
AiBO+ has the following capabilities at the moment:
- Basic postures: sit, stand, lie.
- Locomotion (walking, turning).
- A new, compatible Skitter is bundled. (thanks to DogsBody)
- Avoiding near objects or deepness (e.g a stairway).
- Basic emotions: happy, angry.
- Recognizing the floor surface while walking.
- Crowdsourcing scientific data (machine learning samples).
- Adaptive LED brightness levels.
- Adaptive volume level according to the time of the day.
- Turning over if the robot is upside down.
The AiBO+ AIBOWare must be installed in a PMS stick and an AiBO+ Client application can be run under Windows, Ubuntu Linux and Android.
The installation instructions and a detailed user guide are available on this page:
http://aiboplus.sourceforge.net/userguide.html
A feature video can be watched:
The same video in 3D:
AiBO+ has the following capabilities at the moment:
- Basic postures: sit, stand, lie.
- Locomotion (walking, turning).
- A new, compatible Skitter is bundled. (thanks to DogsBody)
- Avoiding near objects or deepness (e.g a stairway).
- Basic emotions: happy, angry.
- Recognizing the floor surface while walking.
- Crowdsourcing scientific data (machine learning samples).
- Adaptive LED brightness levels.
- Adaptive volume level according to the time of the day.
- Turning over if the robot is upside down.
The AiBO+ AIBOWare must be installed in a PMS stick and an AiBO+ Client application can be run under Windows, Ubuntu Linux and Android.
The installation instructions and a detailed user guide are available on this page:
http://aiboplus.sourceforge.net/userguide.html
Feliratkozás:
Bejegyzések (Atom)